Skip to main content

Full text of "On_Alexanders_Track_To_The_Indus"

See other formats


MACMILLAN AND CO., LIMITED 

LONDON BOMBAY CALCUTTA MADRAS 
MELBOURNE 

THE MACMILLAN COMPANY 

NEW YORK BOSTON CHICAGO 
DALLAS ' SAN FRANCISCO 

THE MACMILLAN COMPANY 
OF CANADA, LIMITED 

, TORONTO 




RUINED BUDDHIST STf)PA OF TOP-DARA, ABOVE HAIBAT-GRAM. 



ON ALEXANDER'S TRACK 

j* 

TO THE INDUS 

PERSONAL NARRATIVE OF EXPLORATIONS 
ON THE NORTH-WEST FRONTIER OF INDIA 

CARRIED OUT UNDER THE ORDERS 
OF H.M. INDIAN GOVERNMENT BY 

SIR AUREL STEIN, K.C.I.E. 

INDIAN ARCHAEOLOGICAL SURVEY 




WITH NUMEROUS ILLUSTRATIONS AND 
MAPS FROM ORIGINAL SURVEYS 



MACMILLAN AND CO., LIMITED 

ST. MARTIN'S STREET, LONDON 

1929 



TO THE MEMORY OF 
COLONEL SIR HAROLD DEANE, K.C.S.I. 

THE GREAT WARDEN OF THE 

INDIAN NORTH-WEST MARCHES 

THIS RECORD IS INSCRIBED 

IN GRATEFUL REMEMBRANCE 

AND SINCERE ADMIRATION 



PREFACE 

THE explorations described in these pages had for their 
scene a region beyond the administrative border of the 
Indian North-West Frontier not previously accessible to 
Europeans. In the initial chapter of this volume a brief 
account will be found of those recent developments in 'tribal 
polities' which through the rise to power of a strong and 
capable ruler in the person of the Miangul Badshah, now 
'Wall of Swat', brought peace to a land singularly favoured 
by nature but for centuries torn by the discord of man. For 
the enlightened spirit with which he welcomed my visit and for 
the unfailing help and care by which he rendered my travels 
both safe and fruitful, I shall ever cherish deep gratitude. 
But equally grateful I feel to those kind friends on this side 
of the border whose willingly offered support, as recorded in 
the same chapter, made it possible for me to explore that 
fascinating country under the generous auspices of H.M.'s 
Indian Government. 

A kindly Fate, and sympathetic comprehension on the part 
of those who officially dispense it, have enabled me, during 
intervals of my forty-one years' Indian service, to carry 
out explorations over the greater part of Innermost Asia, 
and along the whole of those north-western borderlands of 
India which by their historical past have powerfully at- 
tracted me since my early youth. These travels, devoted to 
antiquarian and geographical research, have taken me from 
westernmost China right through Central Asia and from 
the snowy Pamir ranges down to the desolate coast of the 
Ikhthyophagoi by the Arabian Sea. But nowhere did they 
touch ground so replete with historical interest as in that 
comparatively small area to the west of the Indus which 
Alexander's march of conquest towards India for a brief 
span of time illuminates as it were with the light of a meteor. 



viii PREFACE 

It was the main object of my tour to follow up the track of 
the great Macedonian in this region so far as it is at present 
accessible outside Afghanistan. The classical records of his 
campaign would alone suffice to invest these parts with a 
special human interest. But their history has been so ex- 
ceptionally varied and eventful at other periods also, that 
a rapid review of it seems here justified, be it only to provide 
the right background for what the country reveals to us in 
the life of its present day and in the silent ruins of its past. 

We have grown accustomed to divide the ancient world, 
as some do the modern, between East and West. But in 
many ways India stands apart, separated from either by its 
own ancient civilization, just as it is fenced off geographi- 
cally by the ocean and great mountain ramparts. It is on 
this part of the North-West Frontier, where the main routes 
of trade and migration debouch from the Afghan highlands, 
that India, before modern times, came chiefly into contact 
both with the East and the West. 

Long before Alexander's invasion produced the first 
direct impact of the West on India, the great valleys of 
Peshawar and Swat had seen the descent of conquerors from 
that part of the true East which we know as Iran. The 
victory won in prehistoric times by an invading Aryan chief 
on the banks of the Suvastu, the Swat river, is sung already 
in a hymn of the Rigveda. Gandhara, comprising the 
present Peshawar district with the neighbouring tracts, 
figures among the provinces that the great Darius had 
secured for the Persian empire of the Achaemenidian kings 
of kings. 

Alexander's triumphant invasion passed by, indeed, 
without leaving a trace in Indian literature or tradition. 
But Hellenistic princes from Bactria, which Alexander had 
colonized with Greeks, afterwards ruled on both sides of the 
Indus during a couple of centuries and there kept the door 



PREFACE ix 

open for influences derived from the classical West. It is 
a fascinating chapter in history, though we can study it only 
in the fine Greek-modelled coins of these rulers and in those 
sculptures of Graeco- Buddhist art which the ruined Bud- 
dhist shrines of the Swat and Peshawar valleys have pre- 
served for us. 

Then when the great I ndo- Scythian empire of the Kushan 
dynasty had replaced the small Hellenistic chiefships on both 
sides of the Hindukush and had further extended its sway 
beyond the Indus, it was from this north-western border- 
land that fervent religious propaganda carried the Buddha's 
doctrine, together with Graeco- Buddhist art and Indian 
literary culture, into Central Asia and thence into China. 
This spread of Buddhism right across Asia may well be 
considered India's greatest contribution to the civilization 
of mankind in general. These fair border valleys, dotted 
with sacred Buddhist sites, thus acquired special sanctity 
for monastic communities so far away as the Yellow Sea, 
and attracted the visits of those pious Chinese pilgrims 
whose records now serve to guide us among the ruined 
sanctuaries of Swat, their Udydna, 'the Garden'. 

Without these records we should have scarcely anything 
to lift the darkness that descended on this region during the 
centuries when White Hun and Turkish domination suc- 
ceeded the decay of the Indo-Scythian empire. Declining 
Buddhism gave way to lingering Hindu worship and this 
in turn succumbed about A.D. 1000 to the victorious on- 
slaught of Islam under the great Mahmud of Ghazna. From 
the civilization and art which the Muhammadan conquerors 
of India brought with them out of Iran, itself fertilized long 
before by Hellenistic influences, the border tracts could re- 
ceive but little benefit. They soon became a mere passage 
land tenanted by warlike Pathan tribes from the hills, ever 
ready to dispute the 'Gates of India' to any but the strongest 



x PREFACE 

of the new foreign rulers of Northern India. The once 
flourishing territory in which they had settled lapsed more 
and more into barbarism. The Memoirs of the Emperor 
Babar, the great founder of the Moghul Empire in India, 
have little else to tell of Peshawar and Swat than tales of 
frequent hard fighting with the tribes. 

The advent of Sikh power, under Maharaja Ranjit Singh, 
in the first half of the last century, was but a short-lived 
reaction from the Indian side; across the Indus its hold was 
never more than very precarious. Such as it was, the Sikhs 
were unable to extend it to the Swat valley, where the tribes 
under the spiritual leadership of the famous Akhund of 
Swat, the present ruler's grandfather, maintained an un- 
interrupted independence. 

It was left to the British 'Raj', after the annexation of 
the Panjab, to restore peace and steadily reviving prosperity 
to these border tracts, ravaged by centuries of invasion and 
internal disorder; and it has been the destiny of British arms 
to keep watch and ward here ever since. The help I in- 
variably received, wherever my work took me, from the 
officers who share in the hard task of guarding the Frontier, 
will, like the friendships I was privileged to form among 
them, ever rank with the most cherished recollections of 
my life. It was my good fortune to find the earliest of these 
ever helpful friends in Colonel Sir Harold Deane, that 
lamented great Warden of the Marches, who in due course 
became the first Chief Commissioner of the North-West 
Frontier Province, and to his memory this volume must be 
inscribed. 

It only remains to record my thanks for help received in 
connexion with the present publication. They are due to 
the Government of India for their kind permission to make 
this account of my tour in Swat accessible to a wider public 
and to illustrate it by a selection from the photographs 



PREFACE xi 

I took in the course of it. To the Survey of India Depart- 
ment I am indebted for the use of the topographical materials 
secured with the aid of Surveyor Torabaz Khan, who had 
been deputed by it to accompany me on the journey. The 
Royal Geographical Society has kindly allowed reproduc- 
tion of the sketch-map prepared from the original surveys 
and first published in its Journal, while the more detailed 
map of the Pir-sar area, where I believe I have located 
Alexander's Aornos, was drawn and printed under the 
friendly care of Colonel H. T. Morshead at the Geodetic 
Survey Office, Dehra Dun. Here I may conveniently also 
note that the translation of passages in Arrian's Anabasis 
relating to Alexander's campaign between the Panjkora and 
Indus has been taken from Mr. M c Crindle's Invasion of 
India, with such modifications as examination of the Greek 
text appeared to me to render desirable. 

Finally, I must offer my special thanks to the publishers, 
who readily agreed to whatever could make this small 
volume attractive to the eye, and to the Oxford University 
Press, whose care has greatly facilitated its being satisfac- 
torily passed into print in spite of the great distance at 
present between us. On a separate page I have thought it 
useful to name certain publications in which I have recorded 
observations on the early history and antiquities of the 
North- West Frontier gathered in the course of former ex- 
plorations. 

AUREL STEIN. 

CAMP, MOHAND MARG, KASHMIR. 
September 24, 1928. 



Detailed Report on an Archaeological Tour with the Buner Field 
Force. Lahore, Punjab Government Press, 1898. [Reprinted in 
the Indian Antiquary, Bombay, January-March, 1899.] 

Report of Archaeological Survey Work in the North- West Frontier 
Province and Baluchistan, 1904-5. Peshawar, Government 
Press, N.W. Frontier Province, 1905. 

Ruins of Desert Cathay. Personal Narrative of Explorations in 
Central Asia and Westernmost China. Volumes I-II. London, 
Macmillan & Co., 1912. 

Annual Report of the Archaeological Survey of India, Frontier 
Circle, 1911-12. Peshawar, Government Press, N.W. Frontier 
Province, 1912. 

Serindia. Detailed Report of Explorations in Central Asia and 
Westernmost China. Volumes I-V. Oxford, Clarendon Press, 
1921. 

'Alexander's Campaign on the Indian North- West Frontier.' 
London, Geographical Journal, November-December, 1927. 

Innermost Asia. Detailed Report of Explorations in Central Asia, 
Kan-su and Eastern Iran, Volumes I-IV. Oxford, Clarendon 
Press, 1928. 

An Archaeological Tour in Waziristan and Northern Baluchistan. 
'Memoirs' of the Archaeological Survey of India. Calcutta, 
Government of India Press. [In the press.] 



CONTENTS 

PAGE 

I. An Old Transborder Goal I 

II. The Start for Swat .... 9 

III. Visits to Buddhist Ruins .... 17 

IV. Welcome by an Old Friend ... 22 

V. Bir-kot and the Ruins around it . . 30 

VI. Alexander's Invasion of Swat ... 41 

VII. Past King Uttarasena's Stupa ... 49 

VIII. U de-gram and its ancient Fastness . . 53 

IX. At the Badshah's Capital ... 62 

X. Buddhist Remains about Saidu and Man- 

glawar ...... 72 

XI. On the Way to the Swat Kohistan . . 81 

XII. The Entry into Torwal .... 89 

XIII. To the Headwaters of the Swat River . 94 

XIV. Across the Swat- Indus Watershed . . 99 
XV. Over the Shilkai Pass and down Kana . 105 

XVI. The Ascent to Pir-sar . . . .113 

XVII. In Search of Aornos . . . .120 

XVIII. The Survey of Pir-sar .... 128 

XIX. The Story of Alexander's Siege of Aornos . 135 

XX. Aornos located on Pir-sar . . . 143 

XXI. Ancient Remains at Pir-sar and the Name 

of Mount 0$a . . . . . 149 
XXII. Farewell to an Historic Site and its Story . 155 

XXIII. Through Chakesar and Puran . . .160 

XXIV. To Buner and Mount Ham . . .166 
XXV. Departure from Swat .... 172 

INDEX 175 



LIST OF ILLUSTRATIONS 

Ruined Buddhist Stupa of Top-dara, above Haibat-gr&m Frontispiece 

1. View from Political Agent's House, Malakand Fort, towards Lower 

Swat Valley Face page 10 

2. Government House, Peshawar 14 

3. Ruined dwellings and tower on Bandakai Ridge, above Kotah . 14 

4. Relievos from ruined Buddhist shrines of Swat, probably of Nal, 

removed to the Imperial Museum, Calcutta . . . . iS 

5. Ruins of Buddhist Sanctuary, Nal 20 

6. Small shrine at Gumbatuna, above Swift River .... 20 

7. Raja Shah 'Alam, Khushwakt, nephew of the late Chief of Darel 

andTangir 22 

8. Raft crossing Swat River to right bank below Gumbatuna . . 26 

9. Ruined Buddhist Stupa and Sanctuary, Gumbatuna ... 26 
10. Ruins of Buddhist relic towers, south-west of Blr-ko^ . . 30 
n. Ruined Buddhist shrine, Gumbat, Kandag Valley, seen from south- 
west 32 

12. Entrance and Passage of ruined Buddhist shrine, Gumbat, Kandag 

Valley 32 

13. Ruined Buddhist Stupa, Amluk-dara Valley, seen from south-east 34 

14. Ruins of BuddhistStupaandMonastery,T6kar-dara,seen fromsouth 36 

15. Walls of barrage below ruined Buddhist sanctuary, T5kar-dara . 37 

1 6. View up Tokar-dara Glen with ruined Stupa . . . -37 

17. Sketch-map showing ruined stronghold on Bur-kot Hill . . 38 

18. South-eastern portion of fortifications on Bir-kot Hill . . 40 

19. Ruined towers at north-western end of Blr-k6$ Hill ... 46 

20. Walls crowning crest of hill, Rftja Gira's Castle, Ude-gram . . 46 

21. Ruined Buddhist Stupa, ascribed to King Uttarasena, Shankardar 49 

22. Rock face resembling elephant's head, near Ghalagai 50 

23. Rock-carved image of King, in grotto above Stupa, Shankardar . 50 

24. Sketch-map showing ruined stronghold above Ude-gram . . 52 

25. View up slopes of ancient stronghold above Ude-gram . . 54 

26. View down towards Ude-gram and Swat River from crest of Raja 

Gira's Castle . -54 

27. View along fortified hill crest, Raja Gira's Castle, Ude-gram . 56 

28. North-western spur with bastion, Raja Gira's Castle, Ude-gram . 56 

29. Ruined fortifications on easterly spur of Raja Gira's Castle, 

Ude-gram, seen from below spring 58 

30. Wall on easterly spur, Raja Gira's Castle, U$e-gram ... 60 

31 . Ruined tower of outworkguarding spring,Raja Gira's Castle,U$e-grSm 60 

32. Houses of Mingaora, seen from south 62 

33. Towers, office quarters, and dwellings, Saidu .... 62 



LIST OF ILLUSTRATIONS xv 

34. Door of shop with wood-carving, Mingaora . . Face page 64 

35. Hindu trader's shop, G6g-dara 64 

36. Mi&ngul 'Abdul Wahab Gul-shahzada Sahib, Ruler of Swat . 68 

37. Badshah Mi&ngul Gul-shahzada, Ruler of Swat, with son and chief 

attendants 70 

38. Rock-carved relievos of Buddhist divinities, below Sherarai . . 74 

39. Rock-carved images of Bodhisattvas, near Kukrai ... 74 

40. Ruined mounds marking Buddhist Stupas, Sherarai ... 78 

41. Buddhist inscription on rock above Shakhorai .... 78 

42. Ruined Stupa stripped of its masonry facing, near Charbagh . 82 

43. Remains of ruined Stupa at Jurjurai, Janbil Valley ... 82 

44. Crowd of students and others at Garai Madrasah, Chakesar . 84 

45. Jamadar and men-at-arms of escort, from Nikpi-khel Tract . 84 

46. Rock-carved relievo of Bodhisattva, on slope of Nangrial Ridge, 

above Manglawar 86 

47. Stone with the Buddha's miraculous footprints and KharoshthI 

inscription, above Tirat 86 

48. Boulder marking spot of 'Buddha's clothes-washing', on right bank 

of Swat River 88 

49. Mosque and crowd at Churrai, Torwal ..... 88 

50. Wood -carvings on door of house, Churrai ..... 90 

51. Bridge over Swat River at Ain, below Branial . . . -91 

52. Group of Torwalis at Branial 92 

53. Wooden pillars with carvings in loggia of principal Mosque, Branial 93 

54. Loggia with wood-carvings, Yahya Malik's house, Branial . . 94 

55. Lane in Branial, looking up valley 94 

56. View of Asret Valley from above right bank of Swat River . . 95 

57. Northern spur of Koshujan Massif, seen from opposite Airanai . 95 

58. Chodgram Village seen from south : snowy range above Kalam in 

distance 96 

59. Snowy peaks above Jaba Valley seen from above Chodgram . 97 

60. View down the Swat River Valley from above Peshmal . . 98 

61. Panoramic view of head of Swat River Valley, with Kalam ^ 

and mouths of Utrot and Ushu Valleys, from Korunduke , . 
TJ u n- i >i between 

Ridge above Peshmal . 

62. Panoramic view of the snowy range above Mankial, with f \> 

the Koshujan Peaks (18,750 feet) in middle, seen from ' 9 
Korunduke Ridge above Peshmal 

63. Feast after Ramazan at Branial .... Face page 100 

64. Torwall load-carriers collected at Chodgram . . . .100 

65. Roadside halt for tea of Sipah-salar and escort, above Khwaja-khel 10 1 

66. Rock-carved relievo of AvalokiteSvara Bodhisattva, half-buried in 

detritus, near jare Village 101 



xvi LIST OF ILLUSTRATIONS 

67. Newly built fort at Lilaunai Face page 102 

68. Pathan tomb with engraved headstones, below Bilkanai . .102 

69. View from Shilkai Pass towards snowy range at head of Kana Valley 106 

70. Retainers of Khans of Kana gathered at Damdrai . . . 106 

71. Fort of Dost Muhammad Khan of Kana, Bilkanai . . .no 

72. Ghorband River spanned by single rafter, Kar6rai . . .110 

73. Crest of Upal Range seen from Chat 114 

74. Upal Village seen from south-east 114 

75. Snowy range between Gh&rband and Duber Valleys, seen from 

crest of Upal Range 1 16 

76. Plr-sar Ridge seen from south-western slope of Una-sar Peak . 118 

77. Western slopes of Plr-sar seen from Mashlun . . . .118 

78. Burimar alp and slope down to Burimar Gully, seen from Mashlun 120 

79. Mashlun shoulder and Bar-sar cliffs above, seen from below Burimar 1 20 

80. View of snowy range at head of Swat Valley, looking north from 

Lande-sar 126 

81. Una-sar Peak seen from Kuz-sar . . . . . .126 

82. Northern end of Plr-sar Ridge with Bar-sar and Lande-sar above; 

Swat-Indus Watershed Range in distance . . . .128 

83. Cliffs below Kuz-sar end of Plr-sar, seen from Asharai Ridge . 130 

84. Ridges of Dratserge and BenamSz to east of Plr-sar, seen from 

Lande-sar 132 

85. Fields near middle of Plr-sar Ridge, with Bar-sar and Lande-sar in 

distance 150 

86. Remains of walls of ruined fort on top of Bar-sar . . -150 

87. Indus River with snow-covered range towards Kaghan, seen from 

below Kuz-sar 156 

88. Ibrahim Baba and Mir WalT, of Ranzero Hamlet, with other Gujars 

examined on Plr-sar 156 

89. At the 'Middle Mosque' of Chakesar 161 

90. Foot of ruined Stupa in Top-dara, GOkand .... 162 

91. 'Abdul JalU Khan (f), of Chakesar, and Fir5z Khan, of Upal . 162 

92. Chauga Village in Puran 164 

93. View towards middle portion of Buner from Nawe-ghakhe Pass . 166 

94. View down Gdkand Valley from above Sh6dara . . . .168 

95. Crags of main summit of Mount Ham 168 

96. View across Buner from Ramanai Spur 170 

97. Hollow on top of Mount Ham, with sacred spring and Hstian- 

tsang's 'Stone Couches' 170 

MAPS 

Alexander's Campaign on the Indian North-West Frontier at end of book 
Plr-sar and Environs ,, 



CHAPTER I 

AN OLD TRANSBORDER GOAL 

I SHALL not soon forget the joyful excitement with which, 
early in December 1925 on arriving at Delhi after a long 
and busy stay in England, I found awaiting me a letter 
from Sir Norman Bolton, an old Frontier friend and at 
that time Chief Commissioner of the North-West Frontier 
Province, telling me that a goal which I had for many years 
desired was now at last within my reach. This was the great 
transborder tract of Upper Swat and the adjacent valleys, 
which, by their historic past and the many reported vestiges 
thereof, had attracted me ever since as a young student, 
thirty-eight years before, I first came to work on India's 
ancient soil. 

At that time the turbulent independence of Pathan tribes 
barred the way across the picturesque boldly serrated range 
that divides the great valley drained by the Swat river 
from the open plain of the Peshawar district. The Chitral 
campaign of 1895 had, indeed, opened a route cutting 
through the lower end of the main Swat valley, and in 
the narrow strip of tribal territory thus brought under 
'political' control, the friendly interest of Colonel Sir Harold 
Deane, that lamented Warden of the Marches, had allowed 
me, in the course of rapid tours both before and after the 
great Frontier rising of 1897, to examine ruins of Buddhist 
times. After the latter fanatical upheaval I had had an 
opportunity of seeing parts of Buner, the southernmost 
tract of this region, while accompanying General Sir Bindon 
Blood's Field Force on the short punitive expedition of 
1898. But when the fighting was ended, the fascinating 
ground beyond the administrative British border became 
as much closed as ever to European exploration. 

What drew my eyes so eagerly towards Swat was not 



2 AN OLD TRANSBORDER GOAL CH. i 

merely the fame that this region, the ancient Uddiyana, had 
enjoyed in Buddhist tradition, nor the traces that early wor- 
ship and culture were known to have left there in numerous 
as yet unsurveyed ruins. Nor was it only the wish to find 
myself again on the tracks of those old Buddhist pilgrims 
who travelled from China to the sacred sites of Swat, and 
whose footsteps I have had the good fortune to follow in 
the course of my expeditions through the desert wastes of 
Innermost Asia and across the high ranges of the Pamirs 
and Hindukush. May the sacred spirit of old Hsiian-tsang, 
the most famous of those pilgrims and my adopted 'Chinese 
patron saint', forgive the confession : what attracted me to 
Swat far more than such pious memories was the wish to 
trace the scenes of that arduous campaign of Alexander 
which brought the great conqueror from the foot of the 
snowy Hindukush to the Indus, on his way to the trium- 
phant invasion of the Panjab. 

In the autumn of 1904 arrangements made with the neigh- 
bouring tribes by Sir Harold Deane, then my chief, had 
made it possible for me to visit Mount Mahaban, where 
the south-eastern portion of Buner approaches the Indus, 
ground not previously reached by any European. There 
I could survey the height on which, by a conjecture widely 
accepted for half a century, it had been proposed to locate 
the rock stronghold of Aornos, the scene of the most famous 
exploit of that campaign. But a careful examination of the 
topographical features had shown that they could not be 
reconciled with essential details recorded in the Greek 
accounts of that celebrated siege. It was a purely negative 
result, and the state of 'tribal polities' at the time and for 
nearly two decades afterwards precluded any attempt to 
search for the true site of Aornos higher up near the right 
bank of the Indus, in an area to which various considera- 
tions then pointed. 



CH. i ATTEMPTS TO REACH UPPER SWAT 3 

It was not until after my return from my third Central- 
Asian expedition (1913-16), and after calm on the North- 
West Frontier had followed the stress of the war and the 
subsequent Afghan aggression, that I was able to resume my 
attempts to reach this goal. In December 1921 I made a 
rapid tour along the border of the Hazara District where it 
approaches the left bank of the Indus, and tried to gather 
information bearing on a suggestion first made to me by 
my lamented friend Colonel R. A. Wauhope, R.E. Thirty 
years before, on one of the hard-fought Frontier campaigns 
of which the Black Mountains have been the scene, he had 
sighted from afar a high spur descending from the Swat 
watershed to the right bank of the Indus, and there he 
thought that a likely location of Aornos, the 'Rock' Alex- 
ander had captured, might possibly be looked for. But it 
was only by actual exploration on the spot that the sugges- 
tion could be tested, and my hope of securing a chance for 
this was frustrated for several years by the political situa- 
tion, more than usually disturbed, which had then arisen in 
that transborder region. 

The great fertile valley of Swat, now occupied by Pathan 
clans from the point where the great glacier- fed river breaks 
through the alpine gorges of Torwal, can rarely have been 
long free from internecine feuds since the time, about the 
fifteenth century, when its present masters conquered it 
from the original inhabitants of Dard stock. But early in 
the last century a great Muhammadan saint, the famous 
Akhund of Swat, arose in the land. The spiritual authority 
exercised by him, until he passed away at a great age, was 
strong enough to moderate the usual fighting between the 
rival clans and to unite them whenever aggression threat- 
ened, whether from those ruling the plains of Peshawar or 
from the chiefs who for the time being were masters of the 
adjacent territories to the east or west. But since the great 

B 2 



4 AN OLD TRANSBORDER GOAL CH. i 

AkhuncTs death in the seventies, aggravated dissension 
between the several tribal sections of Upper Swat had 
steadily weakened whatever authority was exercised by the 
Mianguls, the descendants of the saint, and the inheritors, 
as guardians of his tomb, of a kind of spiritual supremacy. 

The opportunity offered by this internal division was 
seized by neighbouring hill chiefs to gain control over the 
rich lands of Swat. The ambitious ruler of Dlr, who held 
the valleys between Swat and Chitral, was gradually over- 
running the fertile tracts on the right bank of the river. The 
Nawab of Amb and Darband, independent chief on the 
right bank of the Indus, was invading Buner and threaten- 
ing to absorb the main valley of Swat from the south-east. 
By a lucky chance my visit to Darband on the previously 
mentioned tour of 1921 had allowed me to become ac- 
quainted with the Nawab's son-in-law 'Abdul Jabbar Khan, 
the descendant of a once influential family driven out of 
Swat, just as he was setting out to lead the van of the inroad 
into Buner. A previous attempt of this adventurous young 
man to establish himself as the Nawab's cat's-paw in the 
uppermost tract of Swat had, indeed, ended in failure. But 
it had made him acquainted with that mountain spur higher 
up on the right bank of the Indus in which I was interested, 
and the information he was thus able to give me proved 
useful enough in the end. 

But fortunately for the modern destinies of Swat, and 
incidentally for my own plans of antiquarian exploration, 
the few years following that chance meeting saw the rise to 
power in Swat of a very capable ruler in the person of 
Miangul Gul-shahzada, the elder of the two surviving 
grandsons of the great Akhund. He managed to attach 
firmly to himself some of the more dependable heads of 
clans and to organize a kind of feudal force, provided with 
adequate transport for food-supplies and thus capable of 



CH. i RISE OF MIANGUL RULER 5 

prolonged operations, an unusual thing in Pathan tribal 
warfare. Thus, after hard struggles, in the course of which 
his younger brother was killed, he ultimately succeeded in 
driving out both invaders. 

Having thus become undisputed master of Upper Swat 
he was soon able to extend his sway to Buner, always 
closely linked by tribal relations with Swat, and also to the 
valleys that descend beyond the watershed range towards 
the Indus. It was not long before the Miangul, now sole 
heir to the name, became known to the people by the simple 
designation of Bddshdh or 'ruler'. The new kingdom that 
his energy and sagacity had built up was soon extended to 
its proper geographical limits by the annexation of Torwal, 
the alpine portion of the Swat valley in which the original 
Dard population of the country, though converted to Islam, 
had maintained its independence and distinct language. 

The peaceful consolidation of what had been won by the 
Badshah's successes had since been greatly facilitated by 
the close and friendly relations that he wisely fostered with 
the administration of the North-West Frontier Province. 
But even this fortunate concatenation of events might not 
have sufficed to enable me to realize my long-cherished plan 
of exploration, had not a kindly Fate during those years 
placed the Government's diplomatic relations with the new 
ruler of Swat in charge of my old and ever helpful friend 
Colonel E. H. S. James, then Political Agent for Dir, Swat, 
and Chitral. 

In the summer of 1925 I had written from England to 
Sir Norman Bolton, Chief Commissioner of the North-West 
Frontier Province, submitting my proposal. Acting under 
the instructions kindly given by him, Colonel James suc- 
ceeded, largely through his personal influence with the 
ruler, in obtaining his consent to my visit to his territory 
and to my intended explorations. That I was to be allowed 



6 AN OLD TRANSBORDER GOAL CH. i 

to extend them over the whole of his country instead of the 
comparatively small area to which my original proposal 
had applied made me feel still more grateful for the en- 
lightened spirit in which my request had been met by the 
Badshah. 

Once assured of his generous welcome, I felt confident 
that all necessary assistance would likewise be available 
from the British side, and it was soon forthcoming with a 
promptness that earned my warm gratitude. The Govern- 
ment of India, on the recommendation of Sir John Marshall, 
Director- General of Archaeology, readily sanctioned my 
employment on the proposed tour, together with a grant of 
Rs. 2,000 to meet incidental expenses. Colonel W. J. Keen, 
another valued old friend just then officiating at the head 
of the North-West Frontier administration, greatly en- 
couraged me by his kind personal interest in the enterprise. 
The Survey of India, which had so often helped to make 
my travels geographically fruitful, readily agreed to facili- 
tate the survey of a region that was for the most part 
practically unmapped or very imperfectly known from 
native route reports, by lending the services of one of its 
trained Indian Surveyors, together with all necessary in- 
struments. 

All this contributed to keep my spirits buoyant during 
the few months which had to be allowed to pass before the 
actual start. I knew well that climatic conditions would, 
until towards the close of winter, greatly hamper or alto- 
gether prevent operations on the comparatively high ground 
to which our explorations were to be extended. On the 
other hand, I could not altogether keep my thoughts from 
the risk involved in delay. For who that knows something 
of 'tribal polities' on the Frontier, could ever feel quite 
assured that conditions would remain quiet for some months 
ahead in that volcanic belt beyond the border? ' How often 



CH. i RETURN TO KASHMIR 7 

has it seen the abrupt rise and fall of chiefships, like waves 
suddenly thrown up by a submarine convulsion! 

Fortunately, to Colonel Keen's experienced eye, the 
'political barometer* on that side of the Frontier stood at 
'fair* for the time being, and his encouraging report lessened 
the strain of waiting. Moreover, there was plenty of work 
to keep me occupied during those few months at the capital 
of New Delhi. Patiently I had to wade through hundreds 
and hundreds of large proof pages of Innermost Asia, the 
detailed account of my third Central- Asian expedition, and 
with equal patience to watch the steady progress made in 
the setting up, photographic reproduction, &c., of all the 
mural paintings that I had succeeded in bringing away from 
ruined Buddhist shrines in distant Turkestan. 

But great was the relief when by the middle of February 
I was free to shake the dust of the new Indian capital off my 
feet and of its overabundant office files off my mind in 
order to gain Kashmir, the familiar base of all my archaeo- 
logical enterprises. It was cheering to find myself once 
again at Srinagar, even though the great valley had shed 
all its verdure for wintry bareness, and though those old 
friends, the great surrounding snow-covered mountains, 
were hidden by low clouds and mists during most of my 
stay. After a year and a half spent in 'civilization', whether 
Western or its Indian imitation, many practical prepara- 
tions were needed to get my camp ready for field work. But 
the kindness of my old friend Dr. Ernest Neve, the dis- 
tinguished head of that great institution, the Church Mission- 
ary Society's Hospital, had provided warm and spacious 
quarters, and I had the ready assistance of old retainers. 
So after a fortnight's toil, in which office paperasses still had 
a large share, I set out on March 4th for the Frontier. 

A day's motor drive of nearly two hundred miles by the 
Jhelam Valley Road down to Rawalpindi brought welcome 



8 AN OLD TRANSBORDER GOAL CH. i 

rest and warmth. There, at the great military centre of the 
Panjab, I was met by Torabaz Khan, the hardy Surveyor 
whom Colonel R. H. Phillimore, Director of the Frontier 
Circle of the Survey of India, had kindly helped to select 
for topographical work on my journey. There, too, I was 
able to secure the loan of the modest armament, four army 
revolvers, which had suggested itself as a desirable com- 
plement to our outfit, in view of the tribal milieu I was about 
to enter. The necessary indent order duly applied for at 
Delhi had, through some red-tape misadventure, failed to 
reach the Rawalpindi Arsenal in time. But fortunately a 
kind friend, Major M. A. L. Gompertz, then on the staff 
of the Northern Command, was prepared to act as deus ex 
machina, and the issue was duly obtained. So the eyes of 
my orderly, Shehra Khan, a demobilized veteran from the 
Salt Range, glistened with pride as he moved off with us 
to the station in charge of the precious 'small arms and 
ammunition' to take the train for Peshawar. 



CHAPTER II 

THE START FOR SWAT 

THEY were a delightful three days that I was able to spend at 
the Frontier capital after my arrival from Kashmir. Haunts 
familiar from the years that I had been stationed there 
looked doubly attractive in the glorious sunshine that fol- 
lowed the pouring rain in which I had travelled from 
Rawalpindi. Nor could I have wished to see the bare but 
beautiful hills that surround the great valley from more 
pleasant quarters than those which I enjoyed, thanks to 
Colonel Keen's hospitable welcome to Government House 
(Fig. 2). They were busy days, too; for in succession came 
Afraz-gul Khan, my old travel companion; next Corporal 
'Abdul Ghafur, my new 'handy man' lent from the ist King 
George's Own Sappers and Miners; and lastly Torabaz 
Khan, the Afridi Surveyor. There was plenty to discuss 
and arrange at this little mobilization. It was hard for 
Afraz-gul, who as a youngster had won his spurs on my 
third Central- Asian expedition, to renounce his eager wish 
to join me once more. But he was already under orders to 
start with Major K. Mason on his Karakoram explorations 
and could now help only in so far as his own brave example 
and its rewards would encourager les autres. 

Nor was it easy to find time for the discussion of all the 
points connected with my projected tour, first with Colonel 
Keen, and then also with Mr. H. A. F. Metcalfe, the Politi- 
cal Agent for Swat, Dlr, and Chitral, who was directly con- 
cerned with the preparatory arrangements. Mr. Metcalfe 
had opportunely come down from his post on the Malakand 
partly for 'political' business and partly, I venture to think, 
to let Mrs. Metcalfe benefit by the little distractions of the 
'Peshawar week* just then in progress. After a recent year 
of exile at the Kabul Legation they had both thoroughly 



io THE START FOR SWAT CH. n 

earned the change. To add to the pleasant impressions of 
my short stay, there arrived for one night Sir Francis 
Humphrys, H.B.M.'s Minister at Kabul, and Lady 
Humphrys, both friends of old Frontier times, on their way 
through to the Viceroy at Delhi. 

So it was almost like a rest when, after all my impedi- 
menta had been dispatched and carpets for distant friends 
in England bought and packed, I myself, on the morning 
of March gth, was whirled off in Mr. Metcalfe's comfortable 
car for the Malakand. Right through the width of the big 
Yusufzai plain it took me past Nowshera, now a big Canton- 
ment, and pretty tree-girt Mardan. Those picturesque hills 
of classical form and bareness, rising like rocky islands 
above the plain, greeted me once again as they had in April 
1906, when I was starting on my second expedition to 
Chinese Turkestan. How I wished that I had more time 
to feast my eyes upon them and the verdant expanse of 
fertile land below! Two hours had sufficed to bring us to 
Dargai, where the railway, now developed to full broad 
gauge, aptly ends within the walls and wire entanglements 
of a prim and somewhat bleak fort. 

Then up to the pass the car rushed by that serpentine 
road which I well remembered seeing in the year that it 
was built, immediately after the Malakand had been fought 
for and taken in 1895. But there was a notable change in 
the landscape. In the valley below, once so barren, a new 
river leapt and foamed: the Upper Swat Canal, which, 
brought through the range by a tunnel over two miles long, 
now carries fertility to the eastern half of the Yusufzai 
plain, an emblem of the Pax Britannica. Now that I saw 
this wonderful canal for the first time 'in being', I did not 
wonder that its fame had reached so far away as the foot 
of the T'ien-shan, where in 1908 I had heard honest Turk! 
cultivators inquiring about it with incredulity. 



CH. ii ON THE MALAKAND n 

The Pax Britannica had left fresh marks, too, on the 
once blood-soaked heights of the Malakand (Fig. i). New 
bungalows had been built on the steep slopes and the growth 
of firs and cedars had greatly increased within the area pro- 
tected by 'tribal-proof towers and forts. The Political 
Agent's new house, close to the crest of the spur imme- 
diately overlooking the pass, seemed full of the comforts of 
an up-to-date English home; and Mrs. Metcalfe's care had 
made it delightfully bright with carpets and flowers and 
two splendid specimens of the British Baby. A great 
change it seemed from the days when I had stayed here 
with Sir Harold Deane in the winter following the great 
Swat uprising, and later when Lady Deane was as yet the 
only and wholly unauthorized lady on the Malakand. 
What with final preparations, repairs successfully accom- 
plished to a camera which had shared my bad fall at a wire 
entanglement, and a mail from Europe to attend to, the 
time I was able to pass in that delightfully hospitable home 
seemed far too short, though half a night was added to the 
working day. But the glorious view across the verdant 
Swat valley and towards the distant snow peaks to the 
north-east will ever remain fresh in my memory. 

On the morning of the loth, exactly according to my 
programme, the start was made. Mr. Metcalfe had been 
invited by the Miangul to his 'capital' at Saidu for a day's 
shoot. So it was the chief's motor-car that carried us down 
into the valley and then on to Thana, the biggest village 
within that strip of Swat which since the Chitral campaign 
is 'protected territory', with its tribal population controlled 
by the Political Agent. The fine military road down to the 
bridge crossing the Swat river had quickly followed the 
occupation of 1895, and Chakdara fort, which we passed by 
the river some three miles before reaching Thana, had 
served more than once as my night quarters on my short 



12 THE START FOR SWAT CH. n 

visits to the ruins of Lower Swat fully thirty years before. 
But the change since then in general conditions was very 
manifest. No doubt the men we met on the road, now 
shaded by fine trees, were still going about armed as they 
used to. But no escort of 'Swat Levies' is now needed here 
to protect 'Sahibs', and when the car dropped me at Thana 
I could freely choose my camp in the open some little dis- 
tance from the village. The local Khans, it is true, took 
care to have it properly guarded at night-time. 

The unmetalled road which continues some twenty miles 
farther to the Miangul's 'capital' at Saidu was the latest 
achievement of the ruler whose protection was now to give 
me access to a region that had hitherto been wholly closed 
to Europeans. Mr. Metcalfe met me again on his way 
down forty-eight hours later, pleased with his bag and fully 
satisfied as to the Miangul's friendly disposition towards 
me. I ndeed, nothing could exceed that very capable ' Politi- 
cal's' thoughtful care and kindly interest in my tour. 

It was a real joy to find myself in the peace and freedom 
of my little tent, the true home that I had last enjoyed on 
a favourite high alp of Kashmir fully two and a half years 
before. Pleasant, too, was the feeling that now at last, if 
all went as I wished and hoped, there would be no need for 
that incessant 'rush' and anxious calculation of time which 
had seemed inseparable in the past from all my expeditions 
great and small. But that strenuous work was not lacking 
from the start, an account of the next few days would 
suffice to show, if there were room here to give it in detail. 
I had decided to stop at Thana for two nights in order to 
visit certain ruined sites that I had heard of in the valley 
and hills to the south, and the day following my arrival 
served to show how much hard tramping and climbing were 
necessary if justice was to be done to the interesting remains 
of Buddhist times which abound in this part of Swat. 



CH. ii VISITS OF CHINESE PILGRIMS 13 

Fertile as Swat still is, and thickly populated as it once 
was, the whole of the great valley must have been crowded 
with Buddhist sanctuaries and religious establishments in 
the centuries immediately before and after Christ. This ex- 
plains the care taken by the old Chinese pilgrims to visit 
Swat on their way from the Hindukush to the sacred sites 
of India, and the glowing account that they have left us of 
the land. No doubt, they and other pious visitors knew 
also how to appreciate the material attractions of Swat, the 
abundance and variety of its produce, its temperate climate, 
and the beauty of its scenery. 

These attractions are significantly reflected in the popular 
etymology that has transformed the ancient name of the 
country, Uddiyana, into Sanskrit Udydna, the 'Garden', as it 
meets us in the narrative of old Hsiian-tsang, the most 
famous of those old Chinese travellers. I have had occasion 
to discuss their accounts of Swat in the initial chapter of my 
Serindia, and this is not the place to review them in detail. 
But since we owe to their records almost all that we know 
of Swat during the thousand years or more preceding the 
Muhammadan conquest, and as the memory of them was 
constantly with me on my wanderings, I may as well intro- 
duce their venerable persons here at the outset. 

The earliest of those devoted pilgrims of whom we know 
was Fa-hsien. After crossing amid many hardships the 
deserts of the Tarim basin and the 'Snow Mountains' of the 
Pamirs, he descended from the Hindukush range to Wu- 
ch c ang or Swat about A.D. 403. He had made his way down 
from Darel through those formidable gorges of the Indus 
which no European has so far been able to enter, 'over a 
difficult, precipitous and dangerous road', so his narrative 
tells us, where rock walls forming the sides of mountains 
that rise ten thousand feet above the river had to be passed 
over ladders, 'seven hundred in all'. In Swat he found 



14 THE START FOR SWAT CH. n 

'the religion of Buddha extremely flourishing*. There were 
altogether five hundred monasteries or Sangharamas where 
'wandering mendicant priests were found in everything for 
three days, after which they were told to shift for them- 
selves'. To two of the few sacred spots in Swat that Fa- 
hsien specially mentions, my itinerary will presently take us. 

Sung Yiin, the next pilgrim whose record has survived, 
in A.D. 519 also crossed the Pamirs to the Oxus. But his 
party wisely avoided the dreaded route by the Indus, 'where 
iron chains served for bridges and suspended across the void 
formed a passage', and reached Swat by the way of Kafir- 
istan. Having spent a whole winter and spring in Swat, 
Sung Yiin has left us a full and enthusiastic description of 
the country. He found Buddhism still very flourishing 
there, and the king strictly conforming his conduct to the 
rules of the Buddha's Law. He describes the climate as 
temperate, the soil as fertile, and the people as enjoying an 
abundance of produce. During the night the sound of the 
temple bells filled the whole country. There was a pro- 
fusion of fine flowers, which continued in bloom during 
winter as well as summer. My recollections of the narcissus 
and other early flowers that I had enjoyed on my former 
midwinter visits to Lower Swat, and the loveliness of early 
spring that soon revealed itself to me higher up the valley, 
enabled me to appreciate the truth of this reference. 

Sung Yiin mentions quite a number of spots rendered 
sacred by legends of the Buddha's presence, both in Swat 
proper and Buner. But owing to a certain confusion in the 
sequence of his narrative it might have been difficult to 
locate them correctly did we not possess the account which 
Hsiian-tsang, the greatest of pious Chinese travellers, has 
left us in his Records of the Western Regions. 

When the famous pilgrim arrived about A.D. 630 from the 
side of Kabul he found Buddhism fallen low from its once 




2. (.OVIKNMLNf HOLM., PFSHAWAR. 

of the UiK-i toiiiniissKnifi, Xuith-XVt^t 1 icntni Pro\inco 




CH. ii HSOAN-TSANG'S ACCOUNT OF SWAT 15 

flourishing condition. In Swat, as in the Peshawar valley, 
the ancient Gandhara, the White Hun conquest about a cen- 
tury earlier had left sad traces of ruin. H siian-tsang describes 
quite correctly the varied configuration of Swat, its favour- 
able climate, and the abundance of forest, flowers, and 
fruit-trees. Its people, he tells us, were of a soft and pusil- 
lanimous character, inclined by nature to deceit, and prac- 
tised in magic crafts. Buddhism was still the predominant 
form of worship; but of the fourteen hundred monasteries 
that were said to have once existed by the banks of the 
Su-p'o-su-tu or Swat river, most were in ruins, and the 
number of the brethren whom they contained, once reckoned 
at 18,000, was greatly diminished. Of the monks he sig- 
nificantly records that they were 'fond of reading their texts 
but were incapable of penetrating their meaning, cultivating 
instead the science of magical formulas'. I had previously 
followed the footsteps of my 'Chinese patron saint' through 
the whole width of Central Asia and a great part of Northern 
India, and had then learnt to rely on the general accuracy 
of his topographical indications. In Swat, too, as well as 
in Buner, we might therefore hope to find in him our safest 
guide to the sites once sacred to Buddhist worship. 

The last of the Chinese visitors we know of wasWu-k'ung. 
This humble successor to Hsiian-tsang reached Swat in the 
year 752 in the suite of a mission which the imperial court 
had dispatched to the Turkish ruler of the Kabul valley, 
who then also held Peshawar and Swat. It was the very 
time when China's predominance in Central Asia was about 
to be rudely shaken by Arabs and Tibetans. Wu-k'ung, 
detained by illness in Gandhara, subsequently became a 
Buddhist monk, and after pilgrimages, extending from 
Kashmir to the sacred spots of Bihar, settled down in a 
monastery of the chief place of Swat. 

During his long residence in the country he is said to 



16 THE START FOR SWAT CH. n 

have Visited all the holy vestiges 1 . But his laconic record 
has little more to tell us than that he found 'not the slightest 
difference between what he saw and that which Hsiian- 
tsang's narrative says'. Wu-k c ung's modest brevity is all 
the more regrettable because his return journey from Swat 
to China would have furnished a story of rare interest ; it was 
accomplished under many difficulties and dangers during 
the years 783-90, when the very last strongholds of China's 
Central- Asian power succumbed to the attacks of Turks 
and Tibetans. 



CHAPTER III 

VISITS TO BUDDHIST RUINS 

FROM the scanty records of these pilgrims, distant in time 
and race, I may now turn to the first days spent among the 
silent ruins that alone remain of the Buddhist Swat of their 
age. A long excursion on March nth took me up the wide 
valley to the south through which lead the approaches to 
the passes of Charat and M5ra. Both passes cross the 
watershed towards the Yusufzai plain and are marked by 
remains of ancient bridle-roads, no doubt, of Buddhist 
times. My first visit was to Nal, at the foot of the M5ra 
pass, where, above a small village, diggings made for 
Colonel Deane in 1897 had brought to light a mass of fine 
Graeco- Buddhist relievos. These had been excavated from 
fine Stupas or shrines by local Pathans without proper 
supervision or guidance; but, at least, they were safely 
lodged in the Calcutta Museum. Much regrettable damage 
and loss have been caused, before and since, in tribal terri- 
tory and elsewhere along the Peshawar border, by 'irre- 
sponsible' digging for remains of that Hellenistic sculptural 
art which once adorned all Buddhist sanctuaries of this 
region (Fig. 4). How destructive such digging usually was 
and how often much of the spoil, when sold to amateur collec- 
tors, was ultimately scattered or destroyed, is a story too 
sad to be told here. 

The site near Nal showed grievous signs of such exploita- 
tion (Fig. 5) small fragments of relievos could still be picked 
up among shattered ruins on the surface. But it proved a 
pleasing example of the care with which those old Buddhist 
monks knew how to select sacred spots and place their 
monastic establishments by them. A glorious view down 
the fertile valley to Thana, picturesque rocky spurs around, 
clumps of firs and cedars higher up, and the rare boon of 



18 VISITS TO BUDDHIST RUINS CH. in 

a spring close by all combined to give charm to the spot. 
Even those who do not seek future bliss in Nirvaija could 
fully enjoy it. 

Then up steep rocky slopes, where walls of old monastic 
quarters cling to narrow terraces or nooks in the crags, we 
climbed to the ruins of Kafir-kot, the 'Heathens' Castle'. 
It proved the site of one of those ancient villages which 
evidently had been built for security on difficult ridges and 
hill-tops; their remains abound in these parts. The rough 
but carefully set slabs of stone could, of course, be obtained 
on the spot. Yet the labour involved in constructing the 
terraces that support them, as well as the always massive 
walls of the quarters, must have been immense compared 
with the actual accommodation. How an adequate supply 
of water could be got at such places is somewhat of a 
puzzle unless 'desiccation' supplies a solution. 

This day's work also included the start of our survey 
operations from Dosillo-sar, a high point on the range to 
the east about 4,500 feet above sea-level. It gave us the 
first close view over the nearest portions of the Swat valley 
now ruled by the Miangul. Finally we clambered up and 
down the watershed towards Buner until we reached the 
Mora pass, a narrow gap on the rocky crest. It was a day 
of arduous but instructive toil, and to my relief it showed 
that my two Pathan assistants, Torabaz Khan, of the 
Survey, and 'Abdul Ghafiir, the Sapper Naik, new to my 
ways as they both were, were men of the right stuff. 

On the 1 2th, a day of threatening clouds and occasional 
drizzling rain, I left my camp near Thana to make my 
entry into the Miangul's dominion. As we passed along 
the cart road below the low rocky spur which on its terraced 
slopes bears the densely packed houses of Thana, I could 
see how much this emporium of local trade had grown 
since Lower Swat was firmly brought under British protec- 



CH. in STtJPA OF TOP-DARA 19 

tion. Quite a row of large Sarais and shops, the latter all 
kept by Hindus, now lined the roadside, and the substantial 
mansions of the two chief Khans of the place were to be 
seen above. It was interesting to note that these rambling 
residences, with their arcaded verandahs of timber, lacked 
the defensible towers without which such men of position 
and means, avowed leaders also, no doubt, of local factions, 
would not formerly have considered their safety assured. 
That these and other houses of recent date showed none of 
the elaborate and tasteful wood-carving that adorns old 
and far ruder dwellings of headmen in Upper Swat was 
obviously a penalty paid for the advance of peaceful civi- 
lization. 

From the village of Haibat-gram, a couple of miles 
farther up the main valley, I left the road in a southerly 
direction in order to visit and survey the fairly preserved 
ruin of a Stupa or Buddhist relic tower in the small secluded 
glen of Top-dara. The frontispiece shows this monument of 
Buddhist devotion, erected like all other Stupas of this region 
to enshrine under the solid mass of its masonry some reputed 
relic of the Enlightened One. The protection that the 
massive dome and its bases were intended to afford to the 
bone fragment or other relic of the Buddha had failed, as 
at almost all Stupas of the Frontier, to save the sacred 
deposit from spoliation; for greedy hands, probably long 
ago, had cut through the south-eastern side of the Stupa 
and tunnelled right down the centre to discover and abstract 
what small articles of precious metal, gems or the like, 
might have been placed as a votive deposit with the relic. 

But otherwise the structure had suffered comparatively 
little, as the photograph shows; its distance from the nearest 
inhabited ground had probably saved it from use as a 
quarry. It was accordingly not difficult to determine the 
measures of the lowest base, fifty-two feet by forty-six, and 

ca 



20 VISITS TO BUDDHIST RUINS CH. m 

of the two receding circular bases upon which rested the 
dome proper with its drum over twenty feet high and 
twenty-seven feet in diameter. In some places the shallow 
pilasters still survived which once adorned all the three 
bases prescribed by tradition; but the small stonework of 
which they were built had in general decayed, leaving only 
matrices, as it were, in the solid masonry of large stone 
slabs which formed the facing of the Stiipa. 

A little above the south-eastern side of the quadrangular 
Stiipa base, where the position of the stairs leading up to its 
top was traceable, we found the ruinous remains, overgrown 
by luxuriant vegetation, of what evidently had been a mona- 
stic structure, measuring about a hundred feet square. On 
each side of an open court four small domed chambers 
could be made out, which probably had served as quarters. 
The debris-covered slopes of the gully on either side of the 
ruins may well hide the completely demolished remains of 
small Stupas or shrines such as are found so often at sites 
of this kind. A massive square tower still rising boldly on 
the steep hill-side to the south of the Stupa and some two 
hundred feet above it may well have served as a safe place 
of refuge when danger threatened the establishment. The 
narrow bed passing the foot of the Stupa base is now dry 
except when rain falls on the hill-side behind. But the 
gorge about a quarter of a mile higher up holds a perennial 
spring, and that probably sufficed for the needs of the small 
monastic community. 

As we made our way back to the road north-eastwards 
under threatening rain clouds, we passed scattered ruins of 
massively built structures occupying the crests of small 
detached spurs and extending over more than a mile. The 
character of their masonry showed that they belonged to 
the Buddhist period, like the very numerous dwellings of 
the same type to be found elsewhere in Lower Swat. The 




5 RUINS OK J3UDPH1S1 SANCTUARY, NAL. 




CH. in PAST THE LANDAKAI SPUR 21 

position in which they are always found, commanding 
easily defended slopes, clearly indicates that regard for 
security had been a determining factor in their construction. 
There was no time then for closer examination, and I knew 
that similar ruins of ancient dwellings awaited us farther on. 
Near the village of Jalala we regained what must always 
have been, as it now is, the main line of communication up 
the Swat valley, and following it beyond the village we 
soon approached the Landakai spur. The steep rocky faces 
of this spur descend close to the left bank of the Swat river 
and serve as the natural defence and boundary of Bar- 
Swat or Upper Swat. Its foot during the spring and 
summer months is made quite impassable by the river in 
flood, and the spur itself must often have been held as a 
barrier against invading forces, just as it was at the time 
of the great Pathan rising in 1897. As I rode up the un- 
metalled zigzag road which had recently been built across 
the end of the spur to please the Miangul's desire for modern 
locomotion, I thought how Alexander's Macedonians may 
have toiled up these rock slopes. After this barrier is passed 
the road lies quite open across broad riverine flats and leads 
to the richest and most populous parts of Swat. 



CHAPTER IV 

WELCOME BY AN OLD FRIEND 

WE reached Kotah, the first village of the Miangul's terri- 
tory and a large place surrounded by rice-fields, in drizzling 
rain, and felt glad to find camp duly pitched some distance 
off, outside a small fort of true medieval appearance. Like 
others along the main route up the valley, it had been 
recently constructed, visible evidence of the ruler's correct 
perception of his subjects' still somewhat unsettled alle- 
giance. I was delighted to find myself welcomed there, 
on behalf of the 'Badshah', by Shah c Alam, the nephew 
of Raja Pakhtun Wall and my former guide and protector 
in Darel and Tangir (Fig. 7). I still remembered with 
pleasure how well this attractive young scion of the Khush- 
wakt race, ever alert in body and mind, had looked after our 
safety and comfort when, in 1913, at the start of my third 
Central-Asian journey, I passed through those alpine 
valleys to the north of the Indus which his uncle, Raja 
Pakhtun Wall, had carved into a new kingdom for himself. 
No European had ever visited them before. 

I have had occasion to tell elsewhere the story of the 
brave fights and unscrupulous intrigues, suggestive of 
Renaissance times on Italian soil, by which that capable 
chief had raised himself from the position of a hapless 
refugee from Chitral to that of absolute master of a number 
of turbulent little republics of Dard race. Shah r Alam, 
scarcely as yet emerged from boyhood, but possessed of a 
keen brain and a good native education, was then acting as 
his uncle's Wazlr. Four years after my visit the newly 
created dominion came to a sudden end, by methods similar to 
those which had brought about its rise, through the treacher- 
ous murder of Raja Pakhtun Wall in Tangir. In the course 
of the same day the Raja's main castle at Jalkot had been 




;. RAJA SHAH 'ALAM, KHUSHWAKT, NEPHEW OF THE LATE 
CHIEF OF DAREL AND TANC.lK. 



CH. iv RAJA SHAH 'ALAM OF TANGlR 23 

sacked and burnt, and Shah c Alam obliged to escape with 
his uncle's wives and children across the mountains to the 
valley of Kandia. 

Two plucky attempts he had since made to assert claims 
by inheritance over Tanglr and Darel on behalf of the 
young sons of the murdered ruler; but both had failed owing 
to lack of resources wherewith to secure the adequate sup- 
port of a tribal mercenary force. So Shah 'Alam with the 
two boys had sought refuge with the Miangul, whose power 
might yet help to restore the family's fortune. It was a real 
pleasure to me to see that he had retained his mental and 
physical activity, and as before I found in him a most con- 
genial and helpful companion on all occasions. I had, 
therefore, every reason to feel grateful to the Badshah for 
attaching Shah ' Alam to me. I was glad, too, to observe that 
his present quasi-dependent position did not in any way 
diminish his status with the Swatis, who call him Shahzada 
(prince) or Raja; and it was evident that, like a true Khush- 
wakt, he did not take such ups and downs of fortune too 
much to heart. 

Many a good talk we had together on the long tramps 
that I devoted to ruined 'Gumbats' or domes, i.e. Stupas, 
and other ancient remains during the next few days. There 
was much rain during the first two of them, but this did not 
prevent some interesting surveys. A heavy rain storm 
during most of the night that followed our arrival at Nawe- 
kala, 'the new castle', of Kotah, threatened to turn our 
camping ground into a bog and made me feel very sorry 
for the Badshah's men-at-arms posted as sentries around 
our tents here as at all subsequent camps. Fortunately 
they managed to secure 'charpoys' or strung rope cots on 
which to crouch while keeping their watch. Fortunately, 
too, I could let the Surveyor and Naik seek shelter within 
the fort, dark and uninviting as its quarters looked. The 



24 WELCOME BY AN OLD FRIEND CH. iv 

rain continued to pour through the greater part of the 
morning and made it difficult to attend to anything but 
drainage operations around the soaked tents. Later on, 
however, I was able to set out, under misty skies and 
occasional heavy showers, for the 'old mansions 1 (mdnraz). 
They could be seen crowning a succession of low ridges 
which descend into the broad valley from the east slopes of 
the Landakai spur. 

On all of them we found ruins of ancient dwellings ranged 
one above the other on small terraces along the narrow 
crest of the ridges (Fig. 3). Their characteristic construc- 
tion thick walls of undressed but carefully laid stone slabs 
left no doubt that they went back to the Buddhist period like 
the many similar structures I had previously examined near 
the Malakand and on the slopes of the picturesque hills 
rising like islands above the fertile plain of the Peshawar 
valley, the ancient Gandhara. Almost all these dwellings 
stood well apart from each other, half a dozen or more rooms 
adjoining a square keep-like tower. The latter was in- 
variably built solid up to a height of ten to twelve feet from 
the ground, the entrance having evidently to be reached by 
a wooden ladder or other contrivance, easily withdrawn in 
case of attack. 

This arrangement, as also the very massive masonry of 
the walls enclosing the quarters or small outside courts, 
made it clear enough that defence was the main considera- 
tion. The windows piercing the walls of the rooms 
suggested the same purpose, where they could be traced. 
They were all very narrow on the outside, like loopholes, 
but splayed out within to give the maximum of light 
permitted by such a safeguard. The narrow doorways still 
showed the square holes where heavy wooden cross-bars 
had been fitted on the inside. In spite of the heavy debris 
encumbering most of the interior, it was possible to trace 



CH. iv RUINS OF ANCIENT DWELLINGS 25 

here and there the small neatly walled underground pits 
which had served for the storage of grain. 

There could be no doubt that these defensible structures 
were here and elsewhere intended to provide safety for their 
occupants when danger threatened, whether from outside 
or from local enemies. The construction of such massive 
dwelling-places, difficult of access and far away from culti- 
vable level ground, must have involved an outlay of labour 
incomparably greater than that required to build the rubble- 
and-mud houses that satisfy even the well-to-do among the 
present Pathan population of Swat. It was evident, there- 
fore, that such dwellings could have been built only by 
local headmen and other people of substance. It seemed 
safe, too, to conclude that conditions of insecurity must have 
been frequent during Buddhist times, notwithstanding all 
the pious devotion that prevailed throughout the Swat 
region at that period, as the many ruined sanctuaries and 
the records of our Chinese pilgrims attest. 

On the other hand it would be quite wrong to suppose 
that these defensive towers indicated a particularly bellicose 
character in the population of this tract, as they do at 
the present day on certain parts of the north-western border- 
land, say amongst Afridis or Wazlrs. Against such a mis- 
take I was warned by recollections of the westernmost 
marches of China proper. There, in Kansu, I had seen not 
only every village, but every single outlying farm or hamlet, 
surrounded by high thick walls of stamped clay. But those 
who tried to protect themselves by these defences of quite 
impressive appearance were all peaceable Chinese folk 
and in scarcely any instance had their formidable walls 
helped to save them from the devastating hordes of the last 
great Tungan rebellion. The scarcity of broken pottery 
among these ruined Swat mansions was perhaps a more 
puzzling feature. Was its absence a sign that these defens- 



26 WELCOME BY AN OLD FRIEND CH. iv 

ible structures had served only as places of occasional refuge 
while their owners in ordinary times preferred to dwell 
lower down nearer to water and their lands? Only the 
systematic excavation of more than one such site could 
furnish a definite answer. 

The night brought more heavy rain, and when on the 
following morning it cleared somewhat and a move up the 
valley became possible, it was up a 'high road' resembling 
a wide channel of liquid mud that we tramped. Still, the 
three miles thus covered to near Guratai village were easier 
going than the wide belt of flooded rice-fields between the 
village and the point where the Swat river had to be crossed 
for a visit to ruins reported on the right bank. As through- 
out the main portion of the great valley the river was 
flowing in several branches. The principal one, about a 
hundred and fifty yards wide, was far too deep to be forded 
on horseback, in spite of the early season. So we took to a 
raft of goat-skins, steered by a strong-armed old ferryman, 
which brought us in safety to the opposite bank (F ig. 8) . B ut 
the frail contrivance could carry only five people and was at 
each crossing swept by the swirling river over half a mile 
down stream. So it took some time before our whole party, 
including the escort, was collected and a move made across 
other boggy rice-fields to the small hamlet of Gumbatuna. 

As its name, the 'domes', had suggested, we found there 
a whole group of ruined Stupas nestling between two small 
spurs which an offshoot of the high range to the north sends 
down to the river. This sheltered nook in the hill-side 
immediately overlooks the track that here leads along the 
right bank of the river and must have formed an attractive 
place of pilgrimage for the pious. On an artificially widened 
plateau above the alluvial flat rises a large but badly injured 
Stupa, resembling that of T5p-dara in size and type (Fig. 9). 
By the side of it a high and massive square base is to 




,X KMT CROSSING S\V\1 KIVLR TO RK.HT II \NK BLT O\V GUMBATUXA. 

liii k<~,t lull ii-inu ali-'\t It It batik 




CH. iv SHRINE AT GUMBATDNA 27 

be seen, which may once have carried a shrine or Vihara, 
while the remains of two much broken Stupas of smaller 
size could be traced near the foot of the big one. All these 
structures had been burrowed into a long time ago for 
'treasure* and probably more than once; thus the large 
Stupa, besides having its square base tunnelled into from 
the west, showed a wide shaft sunk down the centre from 
the top. Yet in spite of all the ravages of time and the hand 
of man, parts of the drum and dome still retained layers of 
the hard cement-like plaster that probably once covered the 
outside of all such pious monuments. 

Rain forced us later to seek shelter in the mosque. 
The ancient walled-up terrace on which it stands rises some 
fifty feet above the Stupa plateau and was once, no doubt, 
occupied by monastic buildings, like the neighbouring 
ground on which the hovels of the hamlet are built. Fortu- 
nately a small circular shrine, in a narrow ravine a short 
distance above the mosque and near a fine spring, had fared 
better. The little rotunda, with an interior diameter of 
about fifteen feet, still carries a rather flat dome on its high 
massive walls (Fig. 6). The debris-filled interior may well 
hide remains of the stucco images of Buddhist divinities that 
probably once stood in the shrine. The existence of a spring 
in this picturesque gully and the fine view right across the 
wide valley may have had much to do with the selection of 
this spot by pious tradition as one of the many hallowed by 
the Buddha's visit to ancient Uddiyana. 

Just as I stood by the spring the sun came out at last and 
illuminated a beautiful landscape. Across the wide flood- 
beds of the river to the south-west the rugged hill of Bir-kot 
rose boldly in striking isolation (Fig. 8) ; it was soon to become 
a familiar landmark for us all and for me one of no small 
historical interest. Behind it, far away, showed the snow- 
covered slopes of Mount Ham, dominating from its height 



28 WELCOME BY AN OLD FRIEND CH. iv 

of some 9,200 feet the greater portion of Swat as well as of 
Buner. The head of the fine dome-shaped peak was hidden 
by clouds. Twenty-eight years before I had first sighted 
it from the Buner side. I knew that popular legends even 
now clustered round it, as they did in the old days when 
HsGan-tsang, ever eager in his delightfully naive piety to 
gather wondrous tales, had heard and recorded them. So 
I made obeisance to it from afar, with a firm resolve that 
I should be the first European to tread its heights. And 
then up the river to the west I caught sight of the glittering 
dome of the Shankardar Stupa, a great monument of 
Buddhist worship, of which reports had reached me many 
years before. 

Having recrossed by raft to the left bank we reached the 
foot of the far-stretching Blr-kot hill, whose sheer cliffs drop 
to a deep channel of the river. Scrambling along the rocky 
slopes by a difficult foot-path, we made our way to lush 
meadow land where a lively stream from the south empties 
itself into the river, and then passed the large village of 
Blr-kot. Some distance beyond it I found my camp pitched 
near the spot where the Miangul had started constructing 
a 'Tahsir, fort, jail, and school all of them requisites for 
the consolidation of his rule. Here he had sent his Sipah- 
salar or Commander-in-chief, Ahmad 'All Khan, to await 
me. With his personal bodyguard of some thirty well- 
equipped men, he was to look after me throughout my 
wanderings. And a most pleasant and obliging protector 
this active and keen-witted warrior proved to be (Fig. 37). 

He had fought hard in the Miangul's cause, as more than 
one scar on his lithe body showed. But now that aggression 
from outside had ceased, his chief task was to tour exten- 
sively through his master's dominions, to see that the newly 
built forts at all the important points were properly gar- 
risoned, and in general to show that the ruler, besides a 



CH. iv THE SWAT COMMANDER-IN-CHIEF 29 

quick brain and inherited spiritual authority, was endowed 
also with a sword-arm, ready to enforce his rule. May it 
last long, was my wish, spreading peace in a region greatly 
favoured by nature and already growing in wealth. Some 
day the Badshah's dominion, including the high alpine 
valleys of the Swat Kohistan, may rank as the Kashmir of 
the Frontier, to be admired alike for its natural beauties and 
resources. But for the present I felt quite content to see 
Swat passing happily from turbulent faction to settled con- 
ditions under a strong 'benevolent tyrant'. 



CHAPTER V 

BIR-K5J AND THE RUINS AROUND IT 

village is delightfully situated below the eastern 
end of the hill to which, as we shall see farther on, it owes 
its name. Close to it three large and fertile side valleys 
unite before debouching towards the river. They descend 
from that portion of the watershed range which divides 
Swat from Buner and is crowned by Mount Ham. Their 
upper slopes are well wooded with cedars, firs, and pines, 
and abundance of shrubs and grass clothes the lower slopes, 
at any rate in spring-time. So I enjoyed something like 
alpine surroundings during the days that I had to devote to 
the numerous and important ruins to be found in these 
valleys. My search for them was greatly facilitated by the 
very intelligent and active local guide whom I found in 
young 'Abdul Latif Khan, the son of the chief Khan and 
landowner of Blr-kot; he had joined me as a kind of liaison 
assistant at the suggestion of Mr. Metcalfe. 

'Abdul Latif was the first youth of Upper Swat who had 
received an English education up to the undergraduate 
standard at the Islamia College of Peshawar. He had 
successfully passed the departmental examination for en- 
rolment as candidate for a Naib Tahsildarship, the lowest 
administrative grade in the Frontier Province. But no 
vacancy had as yet offered for his probationary employ- 
ment, and finding time hang heavily on his hands among 
his less enlightened compatriots beyond the border, he 
eagerly accepted the chance of making himself useful to 
a Sahib under the British Government which he hoped 
to serve. Fortunately Raja Shah 'Alam, though English is 
not one of the half-dozen tongues comprised in his linguistic 
equipment, took kindly to this first product of 'Western 
education' among the Badshah's subjects, and this couple 



CH. v RUINS IN KANDAG VALLEY 31 

of devoted companions (Fig. 37) made it easy for me to 
keep my finger, as it were, on local feeling and know- 
ledge. 

I have not room to describe here in detail all the ancient 
remains that four long days of riding and scrambling 
allowed me to survey around Blr-k5t and higher up those 
picturesque valleys. Nor do I propose to record what I 
found and examined in proper chronological order. But 
even brief notes will suffice to show how abundant are the 
proofs of pious ardour and worldly wealth left by the ancient 
Buddhist dwellers in this favoured portion of Swat. When 
I proceeded up Kandag, the westernmost of those valleys, 
I found first a pair of large but ruinous Stiipas within easy 
reach of Bir-kot (Fig. 10). The size of the bases, from sixty 
to seventy feet square, and the conspicuous position they 
occupied made it possible to reconstruct in imagination the 
imposing appearance that these structures once presented. 
But in addition to successive diggings for 'treasure' it was 
clear that quarrying operations for building stones had 
proceeded here for ages, while during more recent times 
search for sculptural remains had helped to complete the 
havoc. 

A couple of miles higher up I found the ruins of a large 
sanctuary with chapels and monastic quarters, known as 
Kanjar-k5te, the 'dancer's mansion', stretching on different 
levels for some hundred and seventy yards, below frowning 
cliffs of red sandstone. Here, too, there was evidence of 
vandal destruction, due to the search for 'Buts', i.e. idols, 
for sale to collectors or dealers in British cantonments. The 
contrast offered by this wild solitude, a small Thebais, to 
the smiling green fields below was strangely impressive in 
the evening. 

An architecturally interesting structure known appro- 
priately as Gumbat, the 'dome' (Figs, n, 12) which rises on 



32 BlR-KOT AND THE RUINS AROUND IT CH. v 

the hill-side near a spring some six miles from Bir-kot, had 
fortunately fared better. It consists of a cella nearly twelve 
feet square within, surrounded on all sides by a narrow 
passage intended for the circumambulation of sacred images 
that Buddhist worship, like Hindu cult to this day, pre- 
scribes. Small windows piercing the massive walls of both 
cella and passage gave light to the interior. This probably 
once contained a colossal standing image of the Buddha; for 
it would be difficult otherwise to account for the great height, 
close on fifty feet, of the dome surmounting the cella. Its 
interior, filled to a considerable height with refuse accumu- 
lations, was occupied by a Gujar family from the neigh- 
bouring hamlet of Balo, while the passage was utilized to 
shelter their buffaloes. On terraces close by I was able to 
trace the position of several small Stupas, now completely 
demolished, and to pick up much-decayed little fragments 
from relievos, which modern searchers for 'idols' had thrown 
aside in the course of their destructive operations. 

Another long day spent in visiting the ruins of a large 
Buddhist shrine high up in the valley of Amluk-dara will 
ever be remembered by me with delight. My visit led me 
up to the very foot of Mount Ham, still invested with 
legendary sanctity, and all the way through charming 
spring scenery. On the way up I passed at Nawagai a 
picturesque village nestling around a fine spring which, as 
ruined walls and terraces in abundance proved, had also 
served as the centre of an ancient settlement. Copper coins 
with Greek legends of Bactrian and Indo-Greek rulers and 
of their I ndo- Scythian successors, including the great 
Kushan kings, were brought to me in numbers as soon as 
the first offerings of this sort had been rewarded with a 
couple of small nickel pieces. They all had been picked up 
on the steep hill-sides above, dotted with massive remains 
of stone-built houses and towers. The chronological evi- 




^ 5 




s 2, 

^ 

x 5 

K 2 



c 

M 
X 



CH. v THE STtJPA OF AMLtJK-DARA 33 

dence that they afforded with regard to a type of ruins so 
abundant throughout Swat proved very valuable. 

Then we left the good mule road, much improved by the 
Badshah, leading up to the Karakar pass, which I had first 
approached in 1898 with a reconnoitring patrol of the Buner 
Field Force. As we turned up a verdant side valley, a 
brisk little stream, fed by the snow-covered heights of 
Mount Ham, refreshed the eye. It made me realize, what 
one is apt to forget on other far more arid parts of the 
Frontier, the benefit conferred by water even without the 
human aid of irrigation. The deep-cut lane along which 
we travelled was lined with rough hedges showing fine 
primrose-like flowers in full bloom, and the trees hanging 
low with their branches, though still bare of leaves, helped 
somehow to recall Devon lanes. Bluebell-like flowers and 
other messengers of spring spread brightness over the little 
terraced fields. At one place they formed what looked like 
a carpet of worship below a large relievo, carved on a 
detached rock and showing a seated Buddha. The pious 
zeal of Pathan invaders Swat was occupied by Yusufzai 
clans only in the fifteenth century has done what without 
too much trouble it could to deface the sacred image. Yet 
this rustic object of worship retained its serenity of head 
and pose. 

Where the little valley bifurcates close above the hamlet 
of Amluk-dara, the steep fir-clad heights of Mount Ham 
came into full view, covered with snow well down from its 
bold pyramid-shaped top. It is sufficiently far from the 
high snowy ranges that rise above the Swat valley farther 
up to offer an impressive sight both from Buner and from 
the middle portion of Swat. It is easy to understand, 
therefore, the veneration which it still enjoys among the 
scattered Hindus of these regions and the superstitious 
legends with which Pathan imagination invests it. We 



34 BIR-KOT AND THE RUINS AROUND IT CH. v 

have here merely another illustration of the rule that no 
conquest nor change of religion can ever efface local 
worship. Who can tell how far back into antiquity this 
sanctity of Mount Ham reaches? 

And just to delight the archaeologist, there rose against 
this grand background a big Stupa (Fig. 13), of carefully 
constructed masonry and in more perfect preservation than 
any I had ever seen. It had not been dug into of old for 
'treasure', like all the Stupas I had so far examined. Nor 
had the 'But* hunter, that destructive agent of modern 
'civilizing' influences, been yet at work among the half- 
dozen decayed mounds marking small Stupas or shrines 
on the terraces immediately behind it. The large monu- 
ment still raises its fine hemispherical dome, about seventy 
feet in diameter, with its stone facing practically intact, to a 
height of some forty-eight feet including the circular drum. 
Together with the customary triple base, the lowest a hun- 
dred and thirteen feet square, the whole structure attains a 
height of close on a hundred feet. 

I can scarcely be mistaken in the belief that, apart from 
small votive Stupas, it is probably the best preserved of 
all the ancient shrines that Indian Buddhist worship has 
raised over supposed relics of its hallowed founder. Nothing 
had fallen but the huge circular stone umbrellas that had 
once belonged to the 'Tee' above the dome. Four of them 
now lay in a heap on the square base of the Stupa. The 
largest of them measures fully fourteen feet in diameter: to 
raise it to that height must have been a task worthy of some 
Egyptian builder. Of two copper coins said to have been 
found at the site one proved to be an issue of the great 
Kushan dynasty and the other of the Turkish Shahis of 
Kabul, thus respectively indicating the approximate prob- 
able dates when the building was constructed and when 
worship ceased. 



CH. v BUDDHIST SITE OF TOKAR-DARA 35 

The survey of this remarkable site, so well chosen by its 
'pious founder', naturally took time, and all the while 
Mount Ham was gathering fleecy clouds round the bold 
crags on its head. Soon after we started back the clouds 
descended from it, and before more than a few miles of the 
return march had been covered a heavy rain storm broke, 
accompanied by much thunder and lightning, and drenched 
us all thoroughly. My escort, however, half a dozen of the 
Badshah's local men-at-arms, were a lusty lot of hardy 
youths and seemed quite cheerful in spite of their wetting. 
When regaled by us with tea in camp they let their thin 
cotton garments dry on their bodies without showing the 
slightest discomfort. 

From Blr-kot I visited another big Buddhist site, quite 
as picturesque as that just described and if anything larger 
(Fig. 14). It lies in a small wooded dale opposite Naji- 
gram village which is known as Tokar-dara. Being more 
easily accessible it has suffered more damage at human 
hands. But this was amply compensated by the interesting 
discovery of an elaborately constructed barrage work, im- 
mediately below the big Stiipa (Fig. 15). It was obviously 
intended to secure a permanent supply of water for what, 
judging from the extensive ruins of monastic quarters, 
must have been a very large community. The spring which 
once may have fed it lies far up on the hill-side (Fig. 16). At 
the same time I found evidence that the reservoir had also 
been planned for the supply of systematic irrigation to the 
terraced fields below. It is the first example so far known 
to me on the Frontier of an ancient engineering work 
designed for this double purpose. Months would probably 
be needed for the complete excavation of the ruins 
higher up; they are thickly overgrown by thorny jungle 
and include a large quadrangle with monastic quar- 
ters. Even confining myself to mere rough survey and 

D 2 



36 BlR-KOT AND THE RUINS AROUND IT en. v 

photographic work, I found it difficult to do it justice in 
a day. 

A very different and to the Western student a much 
higher interest attaches to the ruins of the ancient strong- 
hold that crowns the rugged hill rising above the left bank 
6f the river near the village of Bir-kot. The hill, completely 
isolated, and rising to close on six hundred feet above the 
riverine flat (Fig. 8), forms a very conspicuous landmark 
many miles up and down the valley. It is known as 
the hill of Bir-kot, and it has given its name, meaning the 
'Bir Castle', to the village below. It is here that I was first 
able to identify one of the strong places figuring in the 
Frontier campaign which preceded Alexander's invasion 
of the Panjab and which we know must have taken him 
into Swat. But before we examine the records of Alex- 
ander's historians supporting this location I may briefly 
describe the results of the examination of the Bir-kot hill and 
its remains. 

Where the broad spur flanking the Kandag valley on 
the west approaches the left bank of the river it curves 
round to the north-east. After descending to a low and 
broad saddle crossed by the main road up the Swat valley, 
it rises again, marked along its crest by a succession of bare 
rocky 'kopjes', and ends abruptly in the rugged isolated 
hill of Bir-kot. This hill is washed at its northern foot by 
the river and culminates in a bold rock pinnacle, attaining 
a triangulated height of 3,093 feet, as shown in the sketch- 
plan (Fig. 17). The hill is roughly crescent-shaped and 
drops on its convex side towards the river in precipitous 
rocky slopes, very difficult to climb and in places quite im- 
practicable (Fig. 19). On its concave side, to the south, 
high unscalable crags fringe the central portion of the hill 
and culminate in the rock pinnacle above referred to. 
Towards the south-west the hill runs out in a narrow rocky 




15- WALLS OF BARRAGE BFLOW RUINED BUDDH1S1 SANCIUXRY, 1OKAR-DARA. 




1 6. VIEW UP TOKAR-DARA GLEN WITH RUINED STUPA. 



CH. v RUINED FASTNESS OF BlR-KOT 37 

ridge, utterly bare throughout and for the topmost three 
hundred feet or so of its height very steep. The south- 
eastern extremity of the hill towards Bir-kot village is also 
very steep and terminates in a rocky crest. Where the 
slope affords room for small terraces, these are covered with 
the debris of stone walls marking ancient habitations, and 
with an abundance of potsherds. 

Above the highest of these terraces an imposing stretch 
of wall, massively built of undressed but carefully set stone 
slabs, rises to a height of close on fifty feet (Fig. 18). It pro- 
tected the fortified summit of the hill on that side where the 
natural difficulties of attack were less. At the same time the 
filling up of the space behind it had considerably enlarged the 
level area on the hill-top. This imposing wall continues at 
approximately the same height to the north. It rounds the 
head of a rocky precipitous ravine running down to the 
river, and is thereafter traceable, less massive and less well 
preserved, all along the steep river front. Where the wall 
reaches the north-western end of the fortified summit it 
turns for short stretches to the south-east and south. 

I was here able to trace remains of small towers or 
bastions (Fig. 19) on projecting rocky knolls, intended to 
ward off any attack that might be attempted from the pre- 
viously mentioned narrow ridge to the south-west. Beyond 
this I was able to follow the line of the wall only for a short 
distance; for the hill is at this point faced by sheer cliffs, 
and no defences were needed to make the summit wholly 
unassailable from the plain. Here the rocky pinnacle 
already referred to rises steeply above the level plateau 
formed by the rest of the hill-top. The sides facing towards 
the latter bear remains of ancient masonry wherever there was 
room for walls. Abundant pottery debris strewing the summit 
and slopes made it clear that this commanding position had 
been turned into a kind of citadel and occupied for a long time. 



38 BlR-KOT AND THE RUINS AROUND IT CH. v 

The whole of the circumvallated level area on the top, 
measuring well over two hundred yards in length and more 
than a hundred yards at its greatest width, was found to be 
covered with ruined walls marking decayed habitations. 
A low mound rising above a bastion-like projection at the 
south-eastern end looked as if it hid a small completely 
demolished Stupa. Another at the opposite extremity 
might also have been taken for a ruined Stupa, but for the 
masses of broken pottery that covered it. Most of the 
decorated potsherds picked up among the debris of the site 
could, by the types of their incised or relievo designs, be 
definitely assigned to the Buddhist period. 

In view of the great extent of territory over which my 
surveys were to take me and of the time required for other 
work, I could not attempt systematic excavation either here 
or at any other of the ancient sites that I traced. Nor can 
I detail here the signs of ancient occupation revealed by 
the examination of even the surface remains. But I may 
at least briefly mention the curious relics of ancient means 
of defence that we found while examining the western line 
of wall. We came there upon numbers of round water- 
worn stones of different sizes, undoubtedly brought from 
the river-bed, such as would be used for slings or heavier 
missiles. In one heap, which a little experimental digging 
revealed at a small ruined tower, we discovered no fewer 
than thirty-eight rounds of this antique ammunition. 

An assured water-supply added greatly to the advan- 
tages of Bir-kot as a safe place of defence. So long as the 
hill-top was defended it was practically impossible for an 
enemy to cut off access to the river. A main branch of it 
washes the base of the rocky northern slopes, and the steep- 
ness of the bluffs overhanging the river at this spot shows 
that it must have flowed past them for ages. But there were 
defences on this side of the hill also; for as I descended 



FIG. 17. 

SKETCH-MAP SHOWING 

RUINED STRONGHOLD ON * 

BIR-KOT HILL 
Scale 

100 50 IPO 200 300 400 500 600 700 800 Yards 
uronrcn 1 ~_rjr^-i- . i - --rrr^j^^azrzn 

Conbours, at 40 feet, are approximate 




Del Torabaz Khan. 



CH. v ROCK-CUT PASSAGES TOWARDS RIVER 39 

from the hill-top towards the river, in places with difficulty, 
I noticed remains of old walls and terraces, with abundance 
of ancient pottery everywhere. 

I had been told of two rock-cut passages leading into the 
hill from above the river, and of the local tradition that they 
had served to make access to water still more secure. I was 
shown the entrance to one of them at an elevation of about 
1 80 feet above the river. Once I had passed the low door- 
way, built of ancient masonry of the type peculiar to 
structures of the Buddhist period, I found myself in a 
gallery three feet wide and lined with masonry of the same 
type. At a height of over ten feet it was vaulted with 
horizontal courses of cut slabs. I ascended the gallery for 
only some sixteen yards, to a point where fallen rock partly 
blocked it. Recesses for a square bolt on either side of the 
doorway showed that it could be closed from the inside. 
The exit of another tunnel was found farther to the east and 
nearer to the cliffs overhanging the river. Here a succession 
of natural rock fissures appeared to have been utilized by 
man, for ancient masonry in places lined the rock walls. 
At a distance of some twenty-five yards ascent became 
difficult owing to fallen blocks of stone. Whether the two 
passages meet higher up, as local tradition asserts, could 
be ascertained only by thorough clearing, which would 
take time and adequate preparations. But it appeared to 
me very probable that at least the lower of them, if not both, 
was intended for the purpose above indicated. 

Many coins are found on the top and slopes of the Bir- 
kot hill, especially after rain, and attest the great antiquity 
of the site and its prolonged occupation. Gold or silver 
pieces are melted down promptly or find their way down 
through local Hindus to the coin dealers of Rawalpindi or 
Peshawar. But even so a rapid search made on my behalf 
at Bir-kot village produced a large miscellaneous collection 



40 BlR-KOT AND THE RUINS AROUND IT CH. v 

of pre-Muhammadan copper coins. The specimens ranged 
from issues of the I ndo- Greek and I ndo- Parthian kings and of 
their Indo-Scythian successors, including the great Kushan 
rulers of the first centuries A.D., down to the mintages of 
the 'Hindu Shahi' dynasty which finally succumbed to the 
Muhammadan conquest about the close of the tenth century. 

Coins of the same early periods are also very frequently 
found at the numerous sites marked by the remains of 
Buddhist sanctuaries and ancient habitations around Bir- 
kot and in the side valleys that debouch there. They con- 
clusively prove that Bir-kot must have been the centre of 
a populous and flourishing tract during the centuries which 
immediately preceded and followed the time of Christ. It 
is equally clear that the advantages which the isolated rock- 
girt hill of Bir-kot offered, both by its natural defensive 
strength and its central position on the great highway of 
Swat, were probably appreciated long before the period 
from which the oldest of those coins date. 

Only systematic excavation could reveal how far back 
the earliest occupation of the stronghold dates. But that 
it already existed when Alexander made his triumphant 
advance to the Indus and beyond to the plains of the 
Panjab, I was soon led to conclude from a careful com- 
parison of the topographical and archaeological facts with 
the notices of Alexander's historians regarding his opera- 
tions in this region and the siege of Bazira or Beira in 
particular. 



CHAPTER VI 

ALEXANDER'S INVASION OF SWAT 

HERE we may conveniently turn for a moment from the 
interesting remains of the past which have survived in this 
fastness to a rapid account of what we learn from the 
classical records of the arduous campaign that brought 
Alexander and his hardy Macedonian host into Swat. In 
the spring of 327 B.C. Alexander, after two years of strenu- 
ous fighting in Bactria and Sogdiana, had crossed the 
Hindukush towards Kabul. There he strengthened the 
hold he had previously secured upon this part of what is 
now Afghanistan, and then set out for the conquest of India. 

So far as the country west of the Indus was concerned, 
this meant only a reassertion of the sovereignty of that 
Persian empire of which he claimed to be the heir; for ever 
since the time of the great Darius it had included among 
its satrapies the present North-West Frontier Province, as 
well as most of the Indus valley. But under the last weak 
Achaemenidian 'kings of kings' this sovereignty had prob- 
ably become very shadowy in the mountainous tracts to the 
north. This fact, together with the obvious need of securing 
the flank of his main line of communication, explains why 
Alexander, on arriving in the upper valley of the Kabul 
river, led one corps of his army into the hill country to the 
north, while the rest was to move down into the present 
Peshawar district and secure the passage across the Indus. 

Though we cannot follow the details of Alexander's 
operations by the river Khoes, it is certain that they took 
him for a considerable distance up the large and populous 
valley of the Kunar river. Then he crossed the mountains 
to the east and had more than one hard fight in the territory 
which the Greek records and geographical considerations 
combined clearly show to have been the present Bajaur. 



42 ALEXANDER'S INVASION OF SWAT CH. vi 

The river Guraios, which the Macedonians had to cross 
before Alexander could lead them into the country of the 
Assakenoi, has long ago been proved to be identical with 
the Panjkora, the Gauri of Sanskrit texts. Coming from the 
mountains of Dlr, it flows past Bajaur on the east and then 
joins the Swat river, where it passes through difficult gorges 
towards the Peshawar plain. 

With the passage of the Guraios or PanjkSra, we are told, 
began the invasion of the country of the powerful nation of 
the Assakenoi, and reference to the map shows that this 
could be no other than Swat, as has also been long since 
recognized. The numerical strength of the nation and the 
size of the territory held by it are sufficiently indicated by 
what Arrian records of the army gathered by the Assakenoi 
to oppose his advance. Yet this host, comprising, we are 
told, 'two thousand cavalry and more than thirty thousand 
infantry, besides thirty elephants', did not dare to encounter 
him in the open and dispersed to their several towns in 
order to defend them. 

From this and the account of the several sieges which 
followed it seems safe to infer that the Assakenoi, though 
a brave race, could not have been addicted to those fierce 
ways of fighting which make the present Pathan tribes such 
formidable opponents on their own ground. This con- 
clusion is fully supported by what I have already noted 
regarding the character of the fortified residences found 
scattered on the hill-sides of Swat and the reliance that the 
ancient inhabitants of Swat were evidently accustomed to 
place upon such means of passive defence. It also agrees 
well with the racial characteristics of those people of Dard 
stock and language whom we know to have held Swat 
before the Pathan conquest and remnants of whom I still 
found living now in Torwal, the alpine portion of the 
Badshah's territory. 



CH. vi OPERATIONS AGAINST ASSAKENOI 43 

Arrian and Curtius have left us accounts of the pro- 
tracted operations that were needed for the subjugation of 
the Assakenoi. They show clearly that their territory was 
a large one, extending right away to the Indus. It evi- 
dently included, besides the whole of Swat, Buner and the 
valleys to the north of the latter. These accounts contain 
details of the places besieged and taken by Alexander; but 
it had not been possible to fix their position with any assur- 
ance so long as by far the greater portion of that extensive 
area remained inaccessible to antiquarian research. 

Only for the initial stages of Alexander's march was 
definite guidance available, and that was supplied by plain 
geographical facts. It is certain that in ancient times, as 
at present, the direct route and the only one of any impor- 
tance must have led from the Panjkora through Talash and 
across the easy saddle of Katgala into the wide open valley 
which thence stretches down to the Swat river and its 
strategically important crossing now guarded by the fort 
of Chakdara. Geographical considerations would further 
show us that the several fortified towns which Alexander 
successively besieged and captured were probably situated 
in the main Swat valley; for this at all times must, as now, 
have been the most fertile and populous portion of the 
territory. 

From Arrian, whose account of Alexander's campaign 
is throughout the most reliable and avowedly based on 
contemporary records, we learn that Alexander 'marched 
first to Massaga, which was the greatest city in those parts'. 
Arrian gives a lengthy account of the siege, which ended 
with the city's capitulation after a brave defence of four 
days, after battering engines had been brought up against 
the walls and the chief had been killed. The reference made 
to its chief under the name of Assakenos shows that Massaga 
was considered the capital. Unfortunately Arrian furnishes 



44 ALEXANDER'S INVASION OF SWAT CH. vi 

no clue to the position of the town; nor are we helped in 
locating it by Curtius* rhetorical description of its defences, 
both natural and artificial. 

On general grounds I believe that the site of Massaga 
may probably have to be looked for in Lower Swat. It 
must have comprised then, as now, a particularly large pro- 
portion of rich land, and the ease with which direct access 
can be gained from it by a series of passes to the open plain 
of the Peshawar valley or Gandhara must have greatly 
increased its economic and military importance, as it does 
to-day. It appears very unlikely that Alexander, having 
been brought by his route from the Panjkora straight to 
Lower Swat, could have carried his operations farther up 
the valley before securing his rear and the direct line of 
communication with the rest of his army on the lower Kabul 
river. 

Though the exact location of Massaga cannot be settled 
for the present, I may briefly refer to two significant facts 
connected with its siege. One is the mention made of seven 
thousand Indian mercenaries, brought from a distance, who 
shared in its defence and after its capitulation made a vain 
endeavour to regain their homes and in that attempt were 
exterminated. The employment of so large a paid contin- 
gent from outside clearly indicates command of extensive 
material resources. At the same time it shows that the 
organized defence which the settled population of ancient 
Swat opposed to Alexander was of a very different character 
from that with which a modern invader of tribal Pathan 
territories on the North- West Frontier has to reckon, as 
illustrated by the severe fighting on the Malakand in 1905 
and 1907 or in the memorable Ambela campaign of 1862. 
In the second place it deserves to be noted that, in spite of 
the valour shown by the defenders, Arrian puts the total 
loss suffered by the besiegers at only twenty-five men. 



CH. vi CAPTURE OF MASSAGA 45 

In this cheap price paid for the success we may recognize 
a proof of the ascendancy which Alexander's highly trained 
and war-hardened veterans derived from the possession of 
superior armament; for both Arrian and Curtius bear testi- 
mony to the overmastering effect upon the defenders of 
the besiegers' war engines, including movable towers and 
powerful ballistae. The remarkable transport feat involved 
in carrying this ancient artillery and siege train through the 
mountains, all the way from the Kabul river if not from 
the Hindukush, is a matter well worthy of the attention of 
the modern military student interested in these regions. 

In tracing the further course of Alexander's operations 
in Swat we are fortunately helped by definite archaeological 
and topographical indications. Arrian's Anabasis (iv. 
xxvii) tells us that Alexander 

then dispatched Koinos to Bazira, believing that the inhabitants 
would capitulate on learning of the capture of Massaga. He 
further sent Attalos, Alketas, and Demetrios, the cavalry leader, 
to Ora, another town, with instructions to invest the town until 
he himself arrived. A sally made from the latter place against the 
troops under Alketas was repulsed by the Macedonians without 
difficulty and the inhabitants driven back within their walls. 
With Koinos matters did not fare well at Bazira ; for its people 
trusted to the strength of the position, which was very elevated 
and everywhere strongly fortified, and made no sign of surrender. 
Alexander on learning this set out for Bazira. But having come 
to know that some of the neighbouring barbarians, prompted to 
this by Abisares, were preparing by stealth to enter Ora, he first 
marched to Ora. Koinos was instructed to fortify a strong 
position in front of Bazira, to leave in it a garrison sufficient to 
keep the inhabitants from undisturbed access to their lands, and 
to lead the rest of his force to Alexander. When the people of 
Bazira saw Koinos departing with the greatest portion of his 
troops, they made light of the remaining Macedonians as antagon- 
ists no longer equal to themselves and descended to the plain. 
A sharp encounter ensued in which five hundred barbarians were 
killed and over seventy taken prisoners. The rest fled together 
into the town, and were more strictly than ever debarred from 
access to the land by those in the fortified position. 



46 ALEXANDER'S INVASION OF SWAT CH. vi 

Subsequently, we are told, when the inhabitants learned of 
the fall of Ora, they lost heart and at the dead of night 
abandoned the town. 

In my opinion, the convergent evidence of position, 
remains, and name enables us to locate Bazira with confi- 
dence at the ancient stronghold marked by the ruins on the 
Bir-kot hill. To take the topographical indications first, it 
is clear that Alexander, having made himself master of 
Lower Swat, had necessarily to turn his attention next to 
this stfong place of the 'Blr Castle', which lay quite close 
to what nature has made the great highway up the Swat 
valley. The Bir-kot hill exactly answers the description 
Arrian gives of the position of Bazira 'which was very 
elevated and everywhere strongly fortified*. It is easy to 
understand why, in spite of the impression which the pre- 
ceding capture of Massaga must have produced, no rapid 
success could be gained there by the force under Koinos. 

It is equally clear why Alexander, while himself marching 
upon Ora, situated higher up the valley as we shall see, 
was content, instead of attempting a direct siege of Bazira, 
to leave a small portion of Koinos 1 troops behind for the 
purpose of masking the fastness. Considering its great 
natural strength, nothing less than an arduous and pro- 
tracted siege could hold out promise of success. It was, 
moreover, a position from which it was easy for Alexander's 
opponents to block the main route leading up the Swat 
valley and thus to interfere with any operations that Alex- 
ander might wish to carry out in that direction. Hence the 
order to Koinos 'to fortify a strong position in front of 
Bazira' and 'to leave in it a garrison sufficient to keep the 
inhabitants from undisturbed access to their lands'. Where 
that fortified camp is likely to have stood, it is impossible to 
state with any certainty. Elevated ground near the point 
where the streams from the Kandag and Karakar valleys 



CH. vi BAZIRA LOCATED AT BlR-KOT 47 

meet close above Blr-k6{ village might well have served the 
tactical requirements. 

Turning to the philological side, it is very easy to prove 
that the modern name of Bir-kof, 'the castle of Blr 1 , pre- 
serves in its first part the direct phonetic derivative of the 
ancient local name which the Greek form Bazira was intended 
to transcribe. The Greek letter was regularly used to 
render both the palatal media j and the palatal semi-vowel 
y, two sounds common to the Indo-Aryan and Dardic 
languages, but not known to the Greek alphabet, and vice 
versa. Of this we have conclusive evidence in the Greek 
transcriptions of indigenous names on the coins of rulers 
belonging to this very region and period. Thus two Indo- 
Scythian rulers whose coins are found with exceeding fre- 
quency at sites t>f Swat are known as Azes and Azilizes 
from the Greek legends on their coins, while the Indian 
legends on the reverse call them Aya and Ayilisa. With 
the same ease the gradual phonetic change of the restored 
indigenous form *Bajira or *Bayira into *Baira and then 
into Bir is accounted for by well-known phonetic laws which 
govern the development of all Indo-Aryan languages from 
their ancient into their modern forms and are likewise 
plentifully illustrated in the related Dard languages. The 
addition of the designation kot, 'castle, fort* (Sanskrit 
kotfd) y to the name is readily understood, the term kof being 
in general use throughout the North- West of India and in 
the valleys beyond, whatever the language spoken. 

A striking confirmation of the location of Bazira at Bir- 
ko$ and of the derivation of the modern name of the latter 
is supplied by Curtius. His account, very brief, of the 
operations that followed those in Bajaur tells us that Alex- 
ander 'having crossed the river Khoaspes left Koinos to 
besiege an opulent city the inhabitants called it Beira 
while he himself went on to Mazaga'. There is good reason 



48 ALEXANDER'S INVASION OF SWAT CH. vi 

to believe that in the Khoaspes we have to recognize either 
the Swat river designated by the Iranian form of its ancient 
Sanskrit name, Suvdstu, or else its tributary, the Panjkora. 
Though Curtius, by a manifest error, such as frequently 
occurs in his rhetorical narrative, makes the siege of Beira 
simultaneous with, instead of subsequent to, that of Mazaga 
(Massaga), yet the reference to Koinos makes it certain that 
the Beira he mentions is the same as Arrian's Bazira. In 
this form of the name we have obviously but another 
attempt to reproduce the indigenous *J5ajira or *Bayira. 

Having thus collated the classical records just detailed 
with the results of our survey on and around the hill of 
Bir-kot, I felt encouraged to hope that with the firm footing 
here gained I might now be able to clear up the remaining 
antiquarian questions connected with Alexander's frontier 
campaign. 




21. RUINED BUDDHIST STtJPA, ASCRIBED TO KING UTTARASENA, 

SHANKARDAR. 



CHAPTER VII 
PAST KING UTTARASENA'S STOPA 

ON the morning of March igth, rendered bitterly cold by 
a violent wind which had blown all night, I dispatched my 
camp from Blr-kot village to U de-gram higher up the main 
valley. There was much of interest to observe on the road. 
Only a couple of miles beyond Bir-kot I came to the great 
Stupa of Shankardar (Fig. 21); I had already heard of 
this huge pile in 1897 and had even been able to catch a 
distant glimpse of it through my glasses from the top of the 
Landakai ridge. I found it, alas, in a state of sad ruin. It 
rises at the very foot of a bare rocky hill just where the 
highway skirts it. Probably, in consequence of this, it has 
suffered terrible damage; for the whole village of Shan- 
kardar close by, and perhaps others farther away, had 
utilized the material offered by this convenient quarry. 

All round the two lower bases not only the well-carved 
facing stones but also the greater portion of the interior 
masonry had been removed. Through what remained of 
the lowest base the Badshah's new road had been cut. The 
havoc thus wrought made it impossible to determine the 
dimensions of the ground plan; on the other hand the 
destruction of the bases seemed to increase the impression 
created by the height of the Stupa. From the rough 
measurements taken the diameter of the dome appeared to 
be about sixty-two feet, which is somewhat less than that 
observed at the Stupa of Amluk-dara, and the total height 
over ninety feet from the level of the road. These two large 
Stiipas were very similar in their architectural features 
and in the decorative element introduced by the insertion 
between the fine white facing slabs of little columns of 
black stones. A big cutting made into the dome from the 
north-west showed that treasure-seekers had been tempted 



50 PAST KING UTTARASENA'S STtfPA CH. vn 

to work here long ago. But the mass and solidity of the 
masonry seem to have defeated their, efforts, and the central 
deposit may still be intact. 

A special interest attracted me to this ruined pile; for 
there could be little doubt that it was identical with the 
Stupa whose construction a local tradition recorded by 
Hsiian-tsang attributed to Uttarasena, an ancient king of 
'Wu-chang-na' or Swat. It was believed to cover the share 
which the king had received of the relics of the Buddha's 
body in accordance with the Master's own command preced- 
ing Nirvana. This location of Uttarasena's Stupa is clearly 
indicated, as Colonel Deane recognized long ago, by the 
position which Hsiian-tsang quite correctly assigns to it. 
He describes it as east of the Swat river, at a distance of 
some sixty li or twelve miles south-west of the town of 
Mng-chieh-li, marked, as we shall see, by the present 
Manglawar. 

A little farther on I had the great satisfaction of coming 
upon a striking instance of the accuracy that usually char- 
acterizes the records of my Chinese patron saint. He tells 
us in the 'Memoirs' of his travels of a huge rock, shaped like 
an elephant, close by on the bank of the river. Tradition 
saw in it the body of the white elephant that had brought 
the precious relics for the king; falling dead at this spot it 
had been miraculously turned into stone. It was easy to 
recognize this rock at a point less than half a mile beyond 
the Stupa, where we passed close below cliffs dropping 
precipitously to the road before the village of Ghalagai. A 
projecting part of the rock face (Fig. 22), when seen 
from a short distance, shows a very curious resemblance to 
the head and trunk of an elephant. Here, just as if to prove 
the veneration with which pious eyes used to look up at it, 
I found the head of a roughly cut relievo image, evidently 
that of a Buddha, emerging above rock debris, Muham- 



CH. vii ROCK-CARVED RELIEVO OF KING 51 

madan orthodox zeal had taken care to heap up the latter 
and do serious damage to the rock-cut head. 

An interesting rock-carving some fifty yards farther on 
had fortunately fared somewhat better, owing to its less 
exposed position. There a small natural grotto, some forty 
feet above the road, could be reached with some difficulty 
by clambering over narrow ledges on the steep face of the 
cliff, and in it I was shown a remarkable relievo carved 
from the rock (Fig. 23). The group, about four feet high 
altogether, had also suffered from iconoclast hands. But in 
the middle, on a pedestal supported by lions, I could still 
clearly recognize a bearded figure standing, flanked on 
either side by smaller much damaged relievo images. The 
flame halo rising from the shoulders and the dress of the 
central figure leave no doubt that a royal personage is in- 
tended. The costume comprises a long coat falling over 
bulging trousers stuck into top-boots, and a kind of pelisse 
or mantle hanging from the shoulders. It is of distinct 
interest; for it shows such close resemblance to the dress 
in which the Indo-Scythian rulers of the great Kushan 
dynasty are represented on their coins and rare sculptures 
that the relievo is clearly of approximately contemporary 
origin. Evidently the artist commissioned to raise this 
modest monument to the memory of pious King Uttarasena 
represented him in the habiliments of the rulers of his own 
time, who still retained on the Indian frontier the heavy 
costume brought from their Central- Asian homeland. 

In the course of the next six miles we passed a succession 
of large villages, all nestling along the foot of bare hill spurs, 
while towards the river their fertile alluvial land, all used 
for rice cultivation, stretched in a wide belt. It was easy 
to see from the number of little shops by the roadside (Fig. 
35) that the Hindu traders who almost exclusively keep 
them could make good profits not only as agents for the 

2 



52 PAST KING UTTARASENA'S STtJPA CH. vii 

export of rice to the plains a flourishing trade but also 
out of the active local demand for manufactured goods 
from India. But the total absence of gardens and fruit- 
trees in this fertile and well-watered valley was striking. 

It was a sad illustration, seen also elsewhere in Swat, of 
the effect,? of the surviving Pathan custom of wesh, which 
requires that all land held by a tribal sub-section shall 
change hands among the different families that compose it 
at short intervals, usually of four or five years. Such a 
custom, while significant enough of the democratic spirit 
prevailing among Pathan tribes, is evidently not calculated 
to encourage the planting of trees or gardens by those 
whose tenure of the land would end in a few years. I was 
told that the Badshah was trying hard to wean his subjects 
from this custom, and as the climate of Upper Swat is 
admirably adapted for the growing of fruits of all sorts, 
their sweetness when once tasted by young and old will, 
I hope, facilitate his benevolent efforts. 



FJG. 24. 

SKETCH MAP SHOWING 
RUINED STRONGHOLD 

ABOVE 

DDE-GRAM 
Scale 

TliSuu "J C 70 , ~^ n ^ s j n 6 J 7t MO Yards 

Contoir-. upproximalr or bOfeet intervals 




uv 



w;/i/i/ii /,~~~, * /j 
/*,*, village ' | 

y ^jL-j 

'V' -- 



Rained structures 
Limits of cultivation 
foot path 
Route follow* d 




CHAPTER VIII 

UDE-GRAM AND ITS ANCIENT FASTNESS 

UDE-GRAM, where we halted, proved to be a large place 
with over four hundred households, and pleasant was the 
spot where our tents were pitched between the Badshah's 
fortified 'Tahsll' and the lower edge of a wooded alluvial 
fan. But what prolonged my stay there and made it fruit- 
ful was the unexpected discovery of a large and obviously 
very ancient mountain fastness on the rugged hill range 
that rises above Ude-gram and the neighbouring villages 
on the eastern side of the valley. Imagine a huge ribbed 
scallop-shell turned with its broad edge upwards and its 
narrow mouth resting on gentle alluvial ground : thus, with- 
out sketch-plan (Fig. 24) or photographs, might some idea be 
conveyed of the peculiar hill formation that here had offered 
a natural stronghold in times when there were no fire-arms 
to interfere with safety in a place completely commanding its 
approaches. The site, known to the local Pathans as 'King 
Gira's Castle', was difficult enough to explore even without 
the fear of stones or other missiles from above. It cost us 
two days of stiff climbing along precipitous rock faces and 
along lines of walls that were carried in places over almost 
impossible slopes. But it was well worth the labour and 
fatigue. 

Passing up the alluvial fan above mentioned one enters 
an amphitheatre of steep rocky spurs converging into a 
well-wooded little valley (Figs. 25, 26) and closed at the back 
by the serrated crest of the hill range, fully 2,000 feet above 
Ude-gram. Where this serrated crest, nowhere more than 
about twenty yards wide and in places almost a knife-edge, 
overlooks the fertile side valley of Saidu, nature, for hun- 
dreds of yards, had provided absolutely impregnable 
defences; for the crest falls away on that side in sheer 



54 UPE-GRAM AND ITS ANCIENT FASTNESS CH. vin 

vertical rock walls and in places forms cornices actually 
overhanging them from above (Fig. 20). Where a bare 
narrow spur on that side might bring assailants within 
reach of a stretch of rock wall that bold climbers could 
attack with some chance of success, a strong bastion had 
been built out over projecting cliffs to defeat the attempt. 
But even while keeping to the line of the crest itself (Fig. 27) 
I found progress distinctly difficult over certain stretches of 
this narrow and slippery ground. Yet even over these the line 
of the ancient wall could be traced. One could hardly think 
without distress of the labour involved in its construction, 
at this height and along the difficult rib-like spurs below. 
The amount of human toil and suffering must have been 
out of all proportion to the structural remains that have 
survived the effects of time and of insecure foundations on 
precipitous slopes. 

Along that almost unassailable crest the ancient wall runs 
for close on eight hundred yards. At its southern end it 
overlooks a narrow gully through which, about a thousand 
feet below, a steep footpath passes, connecting Ude-gram 
with the valley of Saidu. Thence the wall turns to the 
north-west and on a precipitous rocky spur descends some 
eight hundred feet to a small and narrow plateau. Defended 
by massive buttressed walls, this plateau projects like a 
bastion and guards what was the most exposed point of the 
fortified area (Fig. 28). From here the circumvallation 
sweeps in an arc round the hollow to the east. It is then 
carried down another two hundred and fifty feet or so, here 
and there over veritable crags, until it strikes a dry torrent 
bed, narrow and rock-lined, down which the main drainage 
of the protected area finds its way to the valley. Here the 
line of wall meets another bold outwork, triangular in shape, 
which defended the approach from this gully. 

Then it runs up on an easterly spur to the other end of 




VIT-.W UP SLOPES Ob ANCIENT STRONGHOLD ABOVE UDE-GRAM. 



CH. viii DEFENCES OF SPRING 55 

the crest which formed the starting-point of our survey (Fig. 
29). In spite of the great steepness of this spur, stretches of 
the wall, with its massive semicircular buttresses, have here 
survived in remarkable preservation (Fig. 30). We found 
it in most places about seven feet thick. The steep slope of 
slippery rock outside this portion of the circumvallation 
must have greatly facilitated defence. Even within the 
wall the slope is so steep that only at a few points was there 
room for quarters occupying small terraces alongside of it; 
all these were found in far advanced decay. Yet their 
masonry had been solid enough, consisting, as in the de- 
fences, of carefully packed layers of stone set in hard 
plaster. 

Where the line of wall curving from the west approaches, 
more than a thousand feet below the crest, the gorge below 
the triangular outwork already mentioned, we found a fine 
perennial spring gushing from among the big boulders that 
fill the bottom of the otherwise dry torrent bed. It was the 
existence of this spring, the only source of water within the 
area, that rendered it capable of being used as a place of 
refuge. There was ample and striking proof of the impor- 
tance attached to the spring in the massive construction of 
the defences intended to guard it. The walls on either side 
descend in double lines into the gorge that holds this 
precious water, and small bastions or towers had been built 
to strengthen the defences wherever any bit of easier slope 
allowed it (Fig. 3 1). Higher up and also towards the south, 
wherever the slopes afforded room, ruined walls of houses 
marked ancient occupation. Their condition of advanced 
decay, as compared with the remains of Buddhist monastic 
quarters, &c., in other places, was clear evidence of their 
great antiquity. Communication over the narrow ledges or 
ladder-like rock paths leading from terrace to terrace must 
at all times have been difficult, and the poor women who 



56 U1?E-GRAM AND ITS ANCIENT FASTNESS CH. vm 

probably had to do most of the water-carrying no doubt 
had an extremely trying task. 

By the side of the spring, ferns and maidenhair were 
growing profusely among the sheltered rock recesses. 
Flowers, too, including true scented violets, found their 
chance of blossoming in the cool humid shade. With one 
old willow-tree already in full leaf, they combined to make 
a picture in pleasant contrast with the grim rocks and 
crumbling fortifications above. It was a spot that might 
well pass for romantic, and I wished that the gifted author 
of Harilek, my friend Major Gompertz, had been with me 
to use it as the scene ready set for one of his clever stories. 

As I descended the gradually widening gorge in the 
evening, I noticed in more than one place carefully walled- 
up terraces that might once have been the site of fruit 
gardens or other places of recreation for the inhabitants of 
that strange stronghold. Now nothing grows on them but 
thickets of thorny Palosa trees. The whole area is held 
sacred to the shrine of holy Pir Khushhal Baba, situated 
farther down by the side of a lively little rill. So none of 
the fertile soil near or below this Ziarat, verdant meadow 
land as it was at this early spring season, may be cultivated. 
I hoped that the time might come when the superior power, 
at once spiritual and secular, of the new ruler of Swat would 
cause it to be planted with those orchards which the valley 
sorely needs and which its sub-alpine climate would dis- 
tinctly favour. 

But how could there be gardens or orchards in a land in 
which, until a few years before, the frequent redistribution 
of all holdings had remained obligatory under the old tribal 
law? Not until the Pax Britannica had slowly but steadily 
made its influence felt up the Swat river, had it been possible 
for this land so favoured by nature to recover from the slow 
decay into barbarism which had already begun when pious 




\ II W MOM, IOKIIIIKI) HILL CKLSI, K \JA (.IRAS (\sTLF, I'DL-GRAM. 




CH. vin RESTING-PLACE OF MUHAMMADAN SAINT 57 

Hsiian-tsang came to visit and describe 'the kingdom of 
Wu-ehang-na, the Garden*. 

Both in the little valley just described and on the alluvial 
fan descending below it (Fig. 26) one could trace other low 
crumbling walls of ancient structures amidst the thick growth 
of scrub and thorny jungle. Such remains were particularly 
numerous at the foot of the spur that bears the south-western 
flanking line of wall. Here a succession of walled-up terraces, 
all once, no doubt, occupied by houses, orchards, or gardens, 
afforded the easiest approach to the fortified area. And this 
explains why the small plateau that forms the end of that 
flanking wall had been turned into a particularly massive 
bastion (Fig. 28), which still rises in places to a height of over 
twenty feet. Over most of the ground near the walls and 
below them broken pottery of distinctly ancient type was to 
be found in plenty. Yet in view of the extreme steepness of 
the slopes above, it seemed to me hard to believe that the 
quarters within the walled area were regularly occupied 
except in times of danger. On the other hand it is improbable 
that the construction of extensive defences on such difficult 
slopes would have been undertaken except for the purpose 
of assuring a safe retreat for the inhabitants of an important 
locality. 

For such a locality the mouth of the valley, where it opens 
towards Ude-gram village, would have afforded ample 
room. It is now covered with extensive Muhammadan 
burial grounds and sacred groves belonging to the Ziarat 
that is here venerated as the resting-place of holy Plr 
Khushhal Baba. Finds brought to me, consisting of the frag- 
ment of a small Graeco- Buddhist relievo, an inscribed Bud- 
dhist seal, and coins of the I ndo- Scythian rulers, pointed 
clearly to the early occupation of this ground. But owing 
to its sacred character systematic search would have been 
difficult, even if time had been available. Pious tradition 



58 UDE-GRAM AND ITS ANCIENT FASTNESS CH. vm 

recognizes in the saintly hero and martyr, Pir Khushhal 
Baba, the leader of the Faithful in the army of Mahmud 
of Ghazna, who after a long siege took 'King Gira's 
fortress' from the last infidel king of Swat. 

Mahmud of Ghazna, the great invader from the Afghan 
highlands, who first laid the north-west of India open to 
conquering Islam, is the oldest historical figure to which 
popular legend on the Frontier reaches back. The tra- 
ditional location of one of his exploits at 'King Gira's 
Castle* would therefore suffice to prove the high antiquity 
that popular belief ascribes to the site. But indications 
furnished by the classical accounts regarding the direction 
of Alexander's operations beyond Bazira, and considera- 
tions connected with the name that I shall presently set 
forth, soon suggested to me the question whether we should 
not look here for the probable location of Ora. 

In order to examine this question we must revert to 
what we have already learned from Arrian about Ora. 
Alexander, after the capture of Massaga, had sent certain 
of his commanders to that town with instructions to invest 
it until he himself arrived. A sally made from the place 
against a portion of the investing force commanded by 
Alketas 'was repulsed by the Macedonians without diffi- 
culty and the inhabitants driven back within their walls'. 
Alexander himself first set out for Bazira, but subsequently 
was induced to proceed straight to Ora on learning of a 
move among the neighbouring barbarians, instigated by 
Abisares, to reinforce its defenders. 

Unfortunately, Arrian's further brief mention of Ora 
supplies no topographical or other local hint. He merely 
tells us that 'he took the town on the first assault against 
its walls and secured the elephants left behind there'. Nor 
does Curtius' account help us. He mentions, indeed, a 
place Nora, to which Alexander dispatched a force under 




29. RUINED FORTIFICATIONS OX EASTERLY SPUR OF RAJA GIRA'S 
CASTLE, HE-GRAM, SEEN FROM BELOW SPRING. 



CH. viii ALEXANDER'S INVESTMENT OF ORA 59 

Polysperchon after the capture of Mazaga, and this has 
been generally assumed to be the same as Arrian's Ora. 
But all that we are told about it is that Polysperchon 'de- 
feated the undisciplined multitude which he encountered 
and pursuing them within their fortifications compelled 
them to surrender the place'. Since the textual records fail 
us, we must feel all the more grateful for the guidance 
afforded by the confident location of Bazira at Bir-k5t. 

That Ora lay higher up the Swat valley than Bazira may 
safely be concluded from two observations. One is the 
reference made to Abisares. We know from his very name, 
the Sanskrit Abhisara, and from other Greek notices in 
connexion with Alexander's further campaign, that this 
chief ruled over the territory on the left bank of the Indus 
where it faces the upper portion of the main Swat valley. 
If Ora was to be reinforced or relieved by tribesmen acting 
under Abisares' instructions or impulse, it was obviously 
because its position farther up the main valley allowed of 
access to it from that side without interference by the Mace- 
donians who had already secured Lower Swat. We are led 
to the same conclusion by the fact that Alexander, as we 
have seen, ordered Koinos, who stood before Bazira, to 
join him for the attack upon Ora with the main portion of 
his force, while taking care to have Bazira masked by the 
remainder. The position occupied by Bir-kot on the main 
line of communication leading up the Swat valley explains 
the necessity of this measure and at the same time clearly 
shows that Ora lay beyond it. 

These considerations, combined with the general geo- 
graphical features, must lead us to look for Ora higher up 
the main Swat valley and at some point which the presence 
of ancient remains would definitely indicate as having been 
occupied by a fortified town of some importance. Now 
Upper Swat above Bir-kot shows, at the present day, a 



60 UPE-GRAM AND ITS ANCIENT FASTNESS CH. vni 

number of large places which might be called towns, such 
as Mingaora, Manglawar, and Charbagh, all on the left 
bank of the river. But at none of these, apart from Ude- 
gram, did I succeed in tracing definite evidence of ancient 
fortification. Nor did I hear of such remains at any of the 
large villages to be found near the right bank. This quasi- 
negative fact would by itself suffice to draw our attention 
to the ruined stronghold above U de-gram as the probable 
position of Ora. But more reliance, I believe, can be 
placed on the evidence which is supplied by the name 
Ude-grdm itself. 

As regards this name it must be explained in the first 
place that it is certainly a compound of which the second 
part is the term gram (Sanskrit grama). Village'. This is 
well known to most Dardic languages and very common in 
old local names of Swat, being attached to the special desig- 
nation just as the word hot, 'fort, castle', is in other names. 
The first part Ude-, also heard as Udi-> is pronounced with 
that cerebral consonant d which to European ears, in 
classical times as now, always sounded like an r and often 
undergoes that change to r in modern Indian and Dardic 
languages. Thus the temptation is great to recognize in 
Arrian's *flpa the Greek rendering of an earlier form of 
this name Ude-, and to derive this name itself from the 
ancient Sanskrit name of Swat, Uddiyana. The phonetic 
changes that such a derivation assumes in the history of 
the name can all be fully accounted for by well-known rules 
affecting the transition of Sanskrit words into later Indo- 
Aryan forms. 

Arrian's account of the impression produced among the 
Assakenoi by the fall of Ora is a proof of the importance 
attaching to the place, and may perhaps also reflect the 
reliance that had previously been placed upon its natural 
strength. We have already seen that the people of Bazira 



CH. vni PROBABLE LOCATION OF ORA 61 

on hearing of the fall of Ora abandoned their town. But 
in addition we learn that 'thus the other barbarians, too, 
did; leaving their towns, they all fled to the rock in that 
country called Aornos'. To Arrian's description of that 
mighty mass of rock and to his account how the fame of 
its impregnability fired Alexander with the ardent desire to 
capture it, I shall recur when relating my subsequent search 
for it by the banks of the Indus. 

In the Swat valley itself, it is clear that the capture of 
Ora had brought Alexander's operations to a triumphant 
conclusion; for Arrian's narrative shows us that, after 
establishing Macedonian posts at Ora and Massaga, as 
well as at Bazira, to guard the country, the conqueror 
turned south to the Peshawar valley. There he was to 
establish his junction with the division of the army that 
had preceded him down the Kabul river, and then to carry 
his campaign farther east to the Indus. 



CHAPTER IX 
AT THE BADSHAH'S CAPITAL 

IT was not the survey of the ruins of 'Raja Gira's Castle* 
alone th^t made my stay at Ude-gram an exceedingly busy 
one. In the close vicinity there were other ancient remains 
to be visited, clearly proving the past importance of the 
place. On skirting the foot of the hill-side towards Gog- 
dara, another large village, scarcely more than a mile away, 
I found a group of Buddhist relievos carved from a rock 
by the roadside, including a colossal seated Buddha; they 
had, of course, all suffered serious damage at the hands of 
iconoclasts. High up in the bare stony gully that runs down 
to Gog-dara a little spring issues from a rock fissure, and 
on a terrace above it rise the ruins of a Buddhist monastery, 
known as the 'quarters of Hasan*. The building, which 
included rows of small vaulted rooms as well as several long 
narrow chambers, had originally been of more than one 
story. But though the top story had fallen, the apartments 
in the two below seemed to need but little repair to make 
them once more habitable for pious mendicants. The fair pre- 
servation of these monastic quarters was in striking contrast 
with the far advanced decay of the buildings within the 
hill fastness above Ude-gram, and thus bore witness to 
the great antiquity of the latter. The Stupa round which 
the devotion of the monkish community and of its pious 
supporters once centred had long ago been wrecked. 

There was also a large mound, indicating prolonged 
occupation in ancient times, to be examined amidst the 
fields close below the present Ude-gram. It appeared to 
be a remnant of an extensive area that had once been in- 
habited and was now probably buried under alluvial de- 
posits or turned into terraced rice-fields. Finds of early 
coins, moreover, were said to be frequent here. But in 



3-2. 11OUSJ.S OF MINdAORX, SJ TN FROM SOI* 11 





TOWKRS, OFFICE QUARTERS, AND DWELLINGS, SAIDU. 



CH. ix VISIT TO MINGAORA 63 

spite of all these vestiges of antiquity I was glad to leave 
U^e-gram on March 23rd and reach Saidu, the hereditary 
seat of the Badshah; for I felt it advisable to meet without 
further delay the remarkable chief whose rise had given me 
access to this fascinating and hitherto forbidden land. 

It was under a brilliantly clear sky that we moved up the 
main valley past Balo-gram and Kambar to where two 
large side valleys descending from the watershed towards 
Buner meet and debouch upon the Swat river. As we 
approached the great bend of the latter, we came into view 
of the great glittering snowy peaks above Mankial, which, 
far away to the north, dominate the uppermost course of 
the river. In the open bay that the valley forms at this 
junction there lay before us Mingaora, now the largest place 
in Upper Swat, quite a little town with its closely packed 
flat-roofed houses (Fig. 32). Fields green with the young 
grass of spring set it off pleasantly against the dark rugged 
spur of Shamelai, which here juts out boldly into the main 
Swat valley and forces the river to change abruptly from 
the southerly course that it has so far maintained and to 
flow almost due west. 

Mingaora, favoured by its situation near the present 
political centre of Upper Swat and of easy access from all 
directions, seemed to be on the way to rivalling Thana as 
a modest commercial emporium. I found the long narrow 
lanes of its Bazar packed not only with people from Upper 
Swat, but also with 'Kohistams' from Torwal and the other 
high valleys on the headwaters of the Swat river. They 
were easily recognizable by their heavy Dard features and 
their coarse woollen cloaks, as were also the rarer and far 
more uncouth figures from Jalkot, Palos, and other turbu- 
lent little republics in the still inaccessible Kohistan of the 
Indus gorges. 

It was interesting to note the quantity of good wood- 



64 AT THE BADSHAH'S CAPITAL CH. ix 

carving displayed on the doors and in the interiors of most 
of the shops (Fig. 34). Among crowded designs of a florid 
geometrical type I recognized graceful acanthus scrolls 
and other floral motifs frequently met with in Graeco- 
Buddhist sculpture. It was welcome proof that the in- 
fluence -of Hellenistic art, to which the relics of ancient 
Swat so strongly bear witness, had not been altogether 
effaced by the flood of barbaric invasion. I was soon to 
find far more distinct traces of it surviving in the domestic 
architecture of Torwal, whose alpine seclusion had helped 
to protect remnants of the old Dard population of Swat. 

In Swat, as elsewhere on the Frontier, there is little hope 
that modern Western influence will serve to revive the 
artistic elements in the crafts of the country. But here, too, 
it must lead to improvements in the material conditions of 
life. And I found welcome evidence of this in the well- 
stocked store newly opened by the Badshah's administra- 
tion. It seemed intended to bring within easy reach of the 
people not only useful agricultural implements, materials 
for the safe transport of produce, and the like, but also 
some civilized luxuries such as books and writing materials. 
A wooden box was hung outside for the reception of letters, 
which at suitable times would be sent on to the British post 
office at Thana, visible proof that contact with the outside 
world was also to be encouraged. 

There were several ruined Stiipas to be visited near Min- 
gaora, all at the foot of the hill-sides overlooking the streams 
which meet there from the side valleys of Saidu and JanbiL 
But none of them were comparable in size to the great 
Stiipas that I had visited near Blr-k5t. Stripped by vandal 
hands of their facing stones to provide convenient building 
material, they had all decayed into shapeless conical mounds. 
So their examination did not take much time, and the 
frowning cliffs on the crest of King Gira's Castle were still 



CH. ix ARRIVAL AT SAIDU 65 

bathed in full sunshine as I rode up the smiling verdant 
valley towards Saidu, the hereditary seat of the Badshah 
and now in course of rapid development into the capital of 
Upper Swat. Here his holy grandfather, the great Akhund 
of Swat, had lived as a spiritual leader. He, too, like so 
many holy men in the west, had known how to choose the 
right spot for pious devotions, while alive, and for local 
worship thereafter. 

Saidu, situated some 3,300 feet above the sea, occupies 
a delightfully open position at the foot of a wooded spur 
descending from an outlier of Ham and dividing two pretty 
side valleys. From afar, as I made my way up, I could 
see the features characteristic of the past and present of the 
site. Amid a cluster of trees and pilgrims' rest houses the 
gilt-domed structure could be seen which shelters the re- 
mains of the holy 'Akhund' or Teacher. Under his spiritual 
leadership Swatis and Yusufzais had for years fiercely 
resisted Sikh aggression, while Maharaja Ranjit Singh, the 
rival 'Buzurg' of the Panjab, was growing old. 

Around, on prominent hill-tops, high towers could be 
seen, quite medieval in appearance (Fig. 33), designed to 
offer safe refuge in case of inter-tribal attack or of sudden 
invasion from the Buner side. The need for them was very 
real in the years when the Mianguls there were two then 
had to carry on a bitter struggle with rivals for secular 
power, whom jealous neighbours, the Nawab of Dir to the 
north-west and the Nawab of Amb, my host in 1921, were 
e &gi n g on and supporting in the interests of their own 
'forward policy'. Now white-terraced mansions of semi- 
European style have risen, since the Badshah has made 
himself sole possessor of his grandfather's sacred inheritance 
and full master of all the land. 

Significantly enough a little higher up, above the new 
residence constructed for the chief's eldest son, stands the 



66 AT THE BADSHAH'S CAPITAL CH. ix 

Badshah's 'Kar-khana' or factory. Here clever artificers 
are at work making fair imitations of Lee-Metfords, quan- 
tities of ammunition, and even small guns. No doubt, the 
Badshah may well have need of support of this kind for 
some time to come. But the Middle School that has been 
established and a well laid out garden which I passed by 
the side of the high road as I came up the valley showed 
that he was not neglecting other elements of stability. 

By the side of the chief's own residence, a rambling place 
with several airy halls and smaller apartments connected 
by screened galleries, I found a large comfortable tent 
pitched for my quarters in a newly planted garden. There 
I was formally welcomed by the Sipah-salar Ahmad 'All 
and his capable elder brother Wazir Hazrat C A1I, the Bad- 
shah's chief executive and judicial officer (Fig. 37). In the 
evening, when duly refreshed, I had a long talk with the 
Badshah in the modest apartment that serves as his private 
council chamber. He assured me at the outset of his full accep- 
tance of my programme. I was to be free to extend my tour 
to any point within his present borders that I might be inter- 
ested to visit. But he insisted on the hospitable condition 
that I and my party should continue to be treated as his 
personal guests, notwithstanding the wish I urged to the 
contrary in view of the probable length of my tour. What 
did half a dozen people more or less matter, he retorted 
with a humorous look in his eye, when he had daily to 
provide for the entertainment of a couple of hundred guests ! 

It was very interesting to listen to the Badshah's lively talk 
about the places that he had visited years ago in the course 
of his Mecca pilgrimage, the holy shrines of Egypt, Pales- 
tine, Syria, and Iraq; for though the spiritual interests 
prompting this voyage were duly emphasized, he had evi- 
dently made it the occasion of many shrewd observations 
on the things of this world. For my own part I had to 



CH. ix TALKS WITH THE RULER 67 

answer questions about the ancient history of Swat. I was 
fortunately able to give him a piece of information that 
interested him more than accounts of Buddhist sanctuaries 
and the like, when I told him that the union of Swat and 
Buner accomplished under his rule had already been duly 
recorded by the early Chinese pilgrims. To satisfy his 
curiosity about my Central-Asian explorations I gave 
Ataullah Khan, his chief Munshi or literatus (Fig. 37), a 
copy of the Persian abstract account of my second expedition. 
It had been prepared years ago in the hope that it might help 
to secure access to the land, alas, still closed for me, of a 
greater ruler beyond the Indian Frontier. 

My first day at Saidu was used, apart from plentiful 
writing work, for a visit prescribed both by interest and 
local etiquette to the shrine of the great Akhund, the 
Badshah's holy grandfather. His tomb rests in a domed 
pavilion, enclosed by open-work screens of which the old 
portions show how much skill and taste still survived fifty 
years ago among the craftsmen of Swat. By their side 
some recent gilded additions looked gaudy. The holy 
Akhund's tomb attracts throughout the year many pilgrims 
from all parts of the Frontier between Peshawar and the 
Hindukush, and the offerings received from them by the 
shrine are justly believed to constitute a very considerable 
annual revenue for the saint's family, the hereditary guard- 
ians of the tomb. Religious students from Swat and differ- 
ent neighbouring hill tracts receive instruction in Islamic 
law as well as bodily sustenance in the 'Jumat' attached to 
the shrine. In the large loggia that serves for their instruc- 
tion I saw some very good wood-carving. I naturally did 
not fail to deposit my own offering to the saint with the 
custodians of the shrine. 

I visited the magnificent spring, shaded by fine old plane- 
trees, that issues close to it, and wondered how much 

F2 



68 AT THE BADSHAH'S CAPITAL CH. ix 

it might have helped to attach the holy teacher to this 
favoured spot during his long life. A visit to the Badshah's 
Darbar hall, a large wooden structure with much good 
carving on columns and panelled walls, concluded the day's 
sightseeing within Saidu. Later in the evening the Bad- 
shah kindly took me in his motor-car down to the river 
bank past Mingaora and showed me the road that he had 
made to facilitate the transport of the heavy timber required 
for his building operations, which is floated down from the 
forests of Torwal. I was glad to find that spiritual obliga- 
tions and political cares had in no way dimmed his keen 
practical perception of the great economic value of the 
magnificent forests of Upper Swat and of the need for their 
systematic protection. 

For two more days my camp had to remain at Saidu 
not on account of the comfortable conditions that we there 
enjoyed, thanks to the hospitable attention of the Badshah 
and his advisers, but because of the many ruins to be 
visited in the neighbouring valleys. So it was only before 
our start in the early morning and on our return in the 
evening from long excursions that I was able to catch 
glimpses of my host and of the daily life surrounding him. 
The Miangul to use the familiar designation inherited from 
his father and still the only one by which he was at that 
time known to the Indian Foreign Department is a person 
of remarkably active habits. The month of Ramazan had 
then just begun. Being strictly observed here as a fast from 
the first sign of dawn to nightfall, it left the Faithful but 
little time for rest and sleep. Yet early as I started in the 
morning, I always found my host about, taking the fresh 
air on the terrace before his Darbar rooms or attending to 
various business. 

Miangul 'Abdul Wahab Gul-shahzada Sahib (Fig. 36) is a 
strongly built person of middle height, with a fine head and 




36. MIANGU1. 'ABDUL WATIAB G1II.-STIAHZADA SAHIB, 
RUI-ER OF SWAT. 



CH. ix PERSONALITY OF SWAT RULER 69 

strikingly intelligent and pleasant features. His age is 
believed to be about forty-five, but owing to the plentiful 
grey hairs in his flowing beard he might well be taken to be 
somewhat older. Like almost all Swatis he is spare in 
figure, and his great fondness for 'Shikar', or game pursued 
on the rugged rocky hills of his country, necessarily keeps 
him 'lean', to use the expression that I often heard Indian 
students apply to friends of old Lahore days who refused 
to grow fat. His quickness of eye and limb impressed me as 
befitting the role he has to play on his newly founded throne. 

That he is quick at the 'up-take', too, I learnt from our 
conversations; for he talks practically nothing but Pashtu, 
and my command of this by no means easy tongue is still 
defective enough to puzzle, as a rule, any new interlocutor 
at first. However, with the Badshah no such trouble 
marred simple conversation. Like many strong rulers in 
these parts he is accustomed to dispose of affairs by word 
of mouth and is apparently little hampered by the fact that 
his knowledge of Persian would scarcely allow him to 
scrutinize closely the actual wording of such documents as 
have to be issued in his name. 

Yet the amount of business that he has to transact as 
supreme arbiter of disputes, as lord of the manor of widely 
scattered landed property, and as chief organizer of his 
revenues and armed forces, must be great; all are tasks that 
demand a record of some sort besides that of a reliable 
memory. The amount of this work could be inferred merely 
from the crowds of Jirgahs (tribal councils), local Khans, 
and other individual applicants whom I saw early in the 
morning already gathered in front of the Miangul's terrace. 
It seemed a very convenient substitute for that takht or 
raised seat of judgement which Indian as well as Central- 
Asian tradition necessarily associates with the function of 
a ruler. 



70 AT THE BADSHAH'S CAPITAL CH. ix 

It was curious, when returning in the evening, to meet 
the same miscellaneous host of Khans, with their followers 
in carefully separated groups, and of humbler folk, walking 
slowly on the wide road that the Miangul has made between 
his seat and Mingaora. They had sought relaxation from 
the long waits and pleadings in watching the game of foot- 
ball played by the schoolboys in a field laid out half-way 
to Mingaora. Now they were returning in the hope that 
the Miangul's flesh-pots would soon enable them to break 
the day's fast. It is a rule enforced by tradition and policy 
that all those who seek justice from him must be entertained 
as his guests, if need be, for three days and not more. A 
wise practice it seems, as calculated both to expedite judge- 
ment and to soften the feelings of disappointed litigants. 
Anyhow, the fiscal expense involved by this system of hos- 
pitality to litigants was obviously in the Miangul's mind 
when, on my request to be allowed to pay for all supplies 
furnished to my party, he smilingly referred to the insig- 
nificance of the charge compared with that of the hundred or 
two of 'guests' whom his kitchen had to entertain daily. I was 
told by others of the reasonable arrangement that divides 
these 'guests' into three distinct classes: one partaking of the 
ruler's own table, another having meals with the Wazir, and 
a third, made up of the common herd, who are served 
direct from the big kitchen. I have little doubt that this 
third division is quite as exacting as the others in respect of 
quantity and not less critical as to the cooking. 

Needless to say that all these parties bringing their cases 
before the Badshah walked about fully armed. The aim of 
even the humblest Pathan cultivator here is to possess a 
rifle, revolver, or pistol, and the variety of weapons met 
with on the road, from the latest magazine rifles and Mauser 
pistols to cheap Martinis, was surprising. Obviously it is 
a good investment to acquire an up-to-date weapon, and as 



CH. ix CLAIMANTS OF JUSTICE AS 'GUESTS' 71 

elsewhere across the Frontier, big prices, up to a thousand 
rupees and more, are paid for the best small-bore magazine 
rifles. The necessity of understanding the sights on these 
has, I imagine, something to do with the knowledge of 
European numeral figures of which I found curious epi- 
graphic evidence, more than once, in graffiti on boulders by 
the roadside. How they may puzzle an archaeologist in 
the distant future! 



CHAPTER X 

BUDDHIST REMAINS ABOUT SAIDU AND MANGLAWAR 

MY visits to ruined Stiipas and other ancient remains in the 
valleys uniting near Saidu revealed not only sites of con- 
siderable interest but also charming mountain scenery. 
These valleys all descend from the boldly serrated and 
abundantly wooded range that divides Swat from Buner. 
The snow-covered pyramid of Mount Ham came again and 
again into view, dominating these sub-alpine landscapes. 
I now began to understand why legends of all sorts, some 
manifestly of very early origin, cluster around this peak and 
cause it to be reverenced not merely by the few local Hindus, 
hardy survivals, as it were, from pre-Muhammadan times, 
but also by Pathans and Gujars. During the previous 
generation or two tribal turbulence and oppression had 
caused the latter to leave Swat for Kashmir with their flocks 
and herds. I was doubly glad to learn that they are now 
being drawn back to the splendid grazing-grounds of Upper 
Swat. It illustrates a beneficent aspect of the peace and 
order that these fair lands derive from the Badshah's new 
rule. Incidentally it also gives hope that the alps and 
forests of Kashmir may suffer less thereafter from over- 
grazing by the Gujars' destructive congeners, their buffaloes, 
goats, and the rest. 

I look back with particular pleasure to the day spent in 
exploring the numerous remains in the large valley that 
descends from the Jaosu pass towards Saidu. It holds a 
number of good-sized villages, all surrounded by fertile 
fields, mostly on terraces. Much of this rich land has either 
descended to the Badshah as church patrimony from his holy 
grandfather or is held by families related to him. Passing 
below the precipitous hill range from which the walls crown- 
ing the crest of King Gira's Castle look down into the verdant 



CH. x ROCK-CUT BODHISATTVA IMAGES 73 

valley, I saw ruins of ancient dwellings on more than one 
low spur. Then above Guligram village we found the large 
Stupa of Shinase in a fair state of preservation, and repeatedly 
came upon rocks by the roadside bearing Buddhist relievos 
(Figs. 38, 39). Almost all of them represented Bodhisattvas, 
in the varamudrd or 'pose of largess'. 

Pious Muhammadan hands had done their best to deface 
the carvings to throw a stone at them as one passes is con- 
sidered a very meritorious act. But there was enough 
detail left to suggest that probably most of these figures 
were meant to represent AvalokiteSvara, pre-eminently the 
dispenser of mercy and help in the northern Buddhist 
Pantheon. I thought of the wide popularity that this Bodhi- 
sattva has enjoyed throughout Central Asia and has re- 
tained to this day as Kuan-yin, the 'Goddess of Mercy', 
in the Far East. I remembered that Hsiian-tsang had 
since his youth vowed fervent worship to AvalokiteSvara, 
and had piously attributed to his divine patron's help his 
escape from death by thirst in the dreaded Gobi at the very 
outset of his great adventurous journey. So I reflected on 
the pleasure with which the pilgrim's eyes must have rested 
on these and other similar rock-carvings. 

Heavy snow still covered Mount Ham. and so my visit 
to it had to be postponed till my return from the Indus two 
months later. But we pushed up the valley as far as Miana. 
From there, at an elevation of about 5,000 feet, a fine view 
offered up to the wooded heights that overlook the pass of 
Jaosu. Spring, with its first flowers, had just arrived at this 
height, and the blossom of the few fruit-trees round the 
hamlet, with the brilliant green of the young shoots of maize 
in the fields, made a delightful foreground. An important 
route to Buner crosses the pass. I was therefore not surprised 
to find that the Badshah's strategic foresight had caused a 
good bridle-road to be made to the pass and had protected 



74 BUDDHIST REMAINS ABOUT SAIDU CH. x 

it by a fort built above the Gujar hamlet of Miana and by 
towers beyond. There might more than once have been 
heavy fighting in this part of the valley. But local tradition 
prefers to ascribe the many crumbling graves that we passed 
in a large sombre grove of wild olive and ilex trees to young 
'children pf the Faith* whom ' white unbelievers' had wickedly 
murdered in the old times. 

There were numerous Buddhist ruins also in the Janbil 
valley through which the route from the Khalel pass leads 
down to Mingaora. There we found nearly a dozen ruinous 
Stupas scattered in little groups; all had suffered badly 
because their accessible position had made them convenient 
quarries of building-stone (Figs. 40, 43). It was probable, 
too, that iconoclastic zeal would assert itself with special 
vigour in a populous neighbourhood. 

By March 2;th I was free to take my leave of the Badshah, 
whose kind consideration for my plans and needs had 
allowed me to use these bright days at Saidu to the full, and 
started on the explorations up the main valley to which I 
had so eagerly looked forward. I was bidden a very cordial 
farewell by the ruler in front of the platform where he dis- 
penses justice to his people. I could read in his keen eyes 
appreciation of the very genuine gratitude I endeavoured 
to express for the generous way in which he had thrown open 
the whole of his dominion to me. Hazrat 'All Khan, the 
Wazir, insisted on accompanying me with a large following 
to the very last houses of Saidu. But even after this, our 
own array had quite an imposing medieval appearance; for 
from here onwards Sipah-salar Ahmad 'All everywhere es- 
corted me, and his selected bodyguard of some thirty well- 
armed and well-set-up 'orderlies' (Fig. 45) greatly swelled 
my usual escort of local men-at-arms. 

The first march was not to be a long one. It was to take 
me to Manglawar, a large place situated at the mouth of a 




;S. ROCK-C^R\ID ki i n:\os or iu*m>HiiT DIMNITILS, BLLOW SHERARAI 




CH. x DEPARTURE FROM SAIDU 75 

side valley to the north of the Shamelai spur, where the 
course of the Swat river makes its sharp turn to the north. 
For a great portion of the year the easier route along the 
main valley is completely closed by the rise of the river, 
which washes the foot of the spur. At such times the track 
leads by the narrow Shamelai pass across the rugged rocky 
heights of the spur. But the spring flood from the melting 
snows in the high mountains had been somewhat retarded 
by the cloudy weather that preceded my arrival at Ude- 
gram, and the winter route was held to be still practicable. 
So we moved down through the Bazar of Mingaora, as 
crowded as before, and a mile beyond reached the point on 
the river bank where precipitous cliffs render the track along 
their foot quite impracticable for animals, even unladen, at 
any time of the year. 

Accordingly, we mounted men had to take to the wide 
bed of the river. Fording its several branches was not alto- 
gether a comfortable affair; for we had left Saidu after 
midday and the volume of water had risen greatly since the 
morning, the swirl of the current being strong enough to 
take men off their feet. Our escort, therefore, wisely kept 
to the footpath along the face of the cliffs. When we had 
safely recrossed a mile or so higher up and joined them 
where the sheer walls of rock recede somewhat from the 
river-bed, I could see how difficult that path is even for 
local foot traffic; for in numerous places it leads over narrow 
rock ledges and gimcrack wooden galleries, not unworthy 
to be compared to those 'Rafaks' which I remembered so 
well in the gorges of Hunza and the uppermost Oxus 
valley. I could well understand the eagerness with which 
the people of Upper Swat now look forward to the time 
when the Badshah shall have carried his high road along 
these forbidding cliffs. It will mean, no doubt, a very tough 
and costly piece of engineering; but for some twenty-six 



76 BUDDHIST REMAINS ABOUT SAIDU CH. x 

miles farther up the valley the extension of the motor road 
would be easy enough and the gain to trade and traffic great. 

Beyond this point we crossed several steep rocky knolls 
by the river where horses had to be led, and at one of them 
I was able to trace the position of an ancient chiusa that 
once guarded the passage. Thereafter, the route lay quite 
open across a fertile alluvial plain to Manglawar. This 
populous village lies at the wide mouth of an important side 
valley which descends from the watershed range towards 
the Indus. It is supposed by local tradition to have been at 
one time the chief place of Upper Swat. In view of its 
position and its name, obviously derived from an earlier 
Sanskrit form Mangala-pura, 'the town of bliss 1 , we can 
scarcely doubt the identity of Manglawar with one of the 
principal towns of Swat in the seventh century A.D., which 
Hsiian-tsang calls Mng-chieh-li, and Wu-k c ung Mang-o- 
p'o. Reports of ancient remains in the vicinity already 
supported this location, and examination of these ruins 
called for a stay of several days there. 

On arrival I found that my protector, the Commander- 
in-chief, whether from a special sense of his responsibility 
for my safety or from due regard for his personal con- 
venience, had had our camp pitched in the centre of the 
large village, on the terraced court in front of the house 
belonging to its principal landowner, Naushirwan Khan. 
The fact that this stands just above the lively little stream 
that descends from the south and divides the village did 
not contribute to its privacy. At first I did not much relish 
the cramped space, the noise made by the large company 
of cheery men-at-arms and the crowd outside, and other 
disadvantages of the site. But when heavy rain started 
that night and poured all through the next day, soaking 
everything outside my little tent and turning the ground 
into a pool of mud, I felt grateful enough for the shelter 



CH. x HALT AT MANGLAWAR 77 

afforded to my people and for the convenient dark-room 
which *Abdul Ghafur, my excellent handy-man, had man- 
aged to improvise in the Khan's 'best room* and armoury 
combined. Advancing 'civilization' had brought here even 
something like a table and two iron chairs, all most welcome 
aids to my work. 

During the next two days and part of the third the 
examination of ancient remains round Manglawar kept me 
busy. Though less numerous and extensive than those to be 
found in the vicinity of Bir-kot, Ude-gram, and Mingaora, 
they afforded abundant proof that Buddhist devotion once 
prevailed here. About half a mile to the east of the northern 
end of the village rises a conical hillock, undoubtedly 
artificial, over sixty feet above the level of the surrounding 
fields. It evidently represents all that survives of that great 
Stupa which Hsiian-tsang describes in a corresponding 
position as marking the sacred spot where the Buddha in 
a previous birth had cut off some of his limbs as a gift 
of charity. The facing slabs of the Stupa and also a good 
deal of the interior masonry had in the course of centuries 
been carried off to serve as building material in the village, 
as local information acknowledged and actual examination 
readily showed. At the foot of the hillock I thought I could 
just trace what seemed like the outlines of the lowest base 
of the Stupa, well over two hundred feet square. Ruinous 
walls of ancient dwellings covered a small spur of the hill- 
side to the north, up to some three hundred feet above the 
river that descends in the valley. These structures, built on 
small terraces, no doubt provided shelter in troubled times 
for inhabitants of the town below. 

Of all the Buddhist relievos found on the rocks. 
Manglawar and miles up the valley (Fig. 46) the^ 
image of a seated Buddha some thirteen feet ir 
certainly the most striking. It is carved on the vjfadl fac 

T 



78 BUDDHIST REMAINS ABOUT SAIDU CH. x 

of a high reddish rock, high above the narrow terrace at its 
foot, and is a well-executed piece of work. This position, 
while it had saved the relievo from damage by pious vandals, 
has at the same time made it distinctly difficult to photo- 
graph. The hands of the true believers must often have 
itched as they saw this benign-faced heathen idol looking 
down towards Manglawar from its height of some three 
hundred feet above Shakhorai hamlet. 

About a mile farther up the valley, but at a lower level 
on the hill-side, there were three Sanskrit inscriptions to be 
visited, excellently incised on huge boulders. Two of them 
are found at a romantic spot where on a precipitous slope 
a big detached rock overhangs a small spring in a grotto 
full of ferns and maidenhair. This cool shady retreat offered 
a delightful contrast to the rugged slopes above and below, 
where every step entangled the explorer in thickets of 
thorny scrub. 

My attempts to make paper squeezes of these two in- 
scriptions were unsuccessful owing to the difficult position 
of the engraved rock surfaces, the strong wind, and lack of 
time. But I managed to secure photographs of them (Fig. 
41) and also of the third inscription, found a few hundred 
yards off on a huge granite boulder known as Khazana- 
gat, the 'rock of the treasure'. I did not, like my escort, 
expect that the inscriptions, engraved in Brahmi characters 
of the first or second century A.D., would reveal to us the 
hiding-places of great treasures. But it was rather a dis- 
appointment to find after my return from Swat that they 
were identical with those which Professor George Biihler, 
that great departed Indologist, had published nearly thirty 
years before from estampages secured by Colonel Deane 
through a clever native agent. . 

What is more regrettable, however, is that these beauti- 
fully engraved and excellently preserved epigraphic records 




40 RUINhl) MOUNDS MARKING BUDDTITST STUPAS, SHURARAI. 




41. BUDDHIST INSCRIPTION ON ROCK ABOVE SHAKHURAI. 



CH. x BUDDHIST ROCK INSCRIPTIONS 79 

yield no historical or antiquarian information. They merely 
reproduce certain famous sayings of the Buddha and brief 
well-known expositions of his doctrine regarding the path 
that leads to salvation, or rather to liberation from the cycle 
of mundane existences. But, no doubt, spiritual merit had 
been gained by the pious donors who had the inscriptions 
engraved, and by those whom they served to guide in the 
right path. 

I am afraid that * Abdul Hanan, the Pathan agent who had 
successfully obtained the estampages for Colonel Deane, 
was not one of the latter; for unable to meet his employer's 
demand for more 'ancient writings' of this sort, he pro- 
ceeded, like that clever rogue, Islam Akhun, whom some 
years later I unmasked at Khotan, to forge them. The 
inscribed stones that he supplied for a couple of years all 
showed, of course, 'unknown scripts'; for the various 'hands' 
that he employed among local Muhammadan theological 
students to make the rude engravings could scarcely be 
expected to learn to imitate the writing of those old Hindu 
infidels. Nevertheless these 'inscriptions' were duly repro- 
duced in learned societies' journals. They might have 
continued to puzzle Orientalist scholars had not the forger, 
emboldened by success, found it more convenient later on 
to supply, instead of engraved stones, impressions that he 
pretended to have taken from inscribed rocks. The fact 
that the supposed originals had in reality been carved on 
wooden planks led in the end to the exposure of the fraud; 
for the impressions were found to reveal not merely the 
mysterious 'unknown characters' but also, only too faith- 
fully, the natural markings of the wood on which they had 
been carved. 

Neither at Manglawar nor elsewhere in Swat were there 
many signs of that charitable regard for the life of animate 
beings which Buddhist doctrine has always and in all lands 



8o BUDDHIST REMAINS ABOUT SAIDU CH. x 

enjoined. We were all the more struck by a curious local 
custom that we here observed. About half a mile to the east 
of the village, alongside of the track leading up the valley, 
there are some large trees where wild-duck abound by day 
and night. They gather there in great numbers from their 
feeding-grounds on the network of channels and pools 
formed by the Swat river and the large stream that joins it 
from the side of Manglawar. The birds while in the trees 
or flying to and from them are considered sacrosanct, 
though elsewhere the Swatis eagerly shoot and trap them. 
No explanation was forthcoming of the asylum thus granted. 
Could it possibly be connected with the fact that the mound 
of the great ruined Stupa already mentioned rises not 
far off? 



CHAPTER XI 

ON THE WAY TO THE SWAT KOHISTlN 

ON the 3ist March I moved my camp to Charbagh, a large 
town-like village situated about three miles farther north 
in the main Swat valley and the centre of much trade. By 
the advantages of its position, in the midst of a wide and 
fertile alluvial plain, Charbagh might well figure among 
the four or five places which at one time or another are 
believed to have ranked as the political centre of old Swat. 
A quantity of Indo-Greek and Indo-Scythian coins brought 
for sale was evidence that the place had long been occupied. 
Ruined Stupas and debris-covered sites of ancient dwellings 
could also be traced at different points of the valley leading 
towards the Kotkai pass, which affords easy access to 
Ghorband and the Indus. 

One of the ruined Stupas rises on the flat bed of the 
valley about a mile to the east. It has no recognizable 
base and now consists merely of a bare core of masonry 
(Fig. 42). I thought I could identify it with Hsiian-tsang's 
miraculous 'Stone Stupa'. This was situated about six miles 
to the north-east of Meng-chieh-li or Manglawar, and was 
believed to have suddenly emerged from the ground where 
the Buddha had stopped to preach the Law. Here, and indeed 
wherever a route connects Swat with the Miangul's newly 
annexed territory to the south and east, a well-constructed 
new road, fit for laden transport of all kinds, attests the 
ruler's grasp of the strategic importance of good com- 
munications. But the corvee for all this must be heavy. 

As I looked up towards the easy Ghorband pass and the 
wide open valley leading down to Charbagh I thought of 
the relief and delight with which the old Chinese pilgrims, 
after travelling from Darel through the difficult gorges of 
the Indus, must have greeted this view, which marked the 

G 



82 ON THE WAY TO THE SWAT KOHISTAN CH. xi 

end of their troubles. Pious Fa-hsien, who made his way 
down the Indus to Swat about A.D. 403, has left us a graphic 
description of this terrifying 'route of the hanging chains' 
with all its dangers and fatigues. The formidable obstacles 
with which nature has beset it have also been recorded 
in the sober pages of the Chinese Imperial Annals. No 
European has ever passed these forbidding gorges of the 
Indus Kohistan. But I was able to hear details about its 
'Rafales' and risks direct from Raja Shah *Alam. Splendid 
cragsman as he is and as nimble with hands and feet as any 
of his Khushwakt race, he spoke with undisguised aversion 
of his seven days' exhausting journey down that succes- 
sion of precipices overhanging the great swirling river in its 
narrow channel. I wish that I might yet accomplish it 
with him! But there is little chance of the Pax Britannica 
opening a way in the near future through these lawless 
Kohistani communities. 

Shah Alam, together with Raja Pakhtun Wall's two 
young sons Firamorz Khan and Jahanglr, is now eating 
the bread of an exile from the Miangul's bounty. He does 
it with innate dignity and with firm hope of a return of 
Fortune's favour. For both of these he claims my genuine 
regard, and, I feel, he appreciates it. Since the downfall of 
his uncle and master, he has twice made an attempt to 
regain Tangir and Darel, in the name of his cousins, but 
in each case it failed after some fighting through want of 
outside support. Whether it could be renewed with some 
chance of success, I could not judge. He had first found 
refuge in the mountain tracts of Kandia, which he knows 
well and has friends in. So I guessed that the support he 
now receives in Swat may be due to reasons not altogether 
altruistic. History in Kashmir and elsewhere records many 
examples of exiled pretenders successfully asserting their 
claims with the help of neighbours. So that there would 



CH. xi MOVE ABOVE CHARBAGH 83 

be nothing new in such an event. I was glad that chance 
had brought me into personal contact with the men who 
seem likely to participate in it. 

From Charbagh I started on April 2nd on my journey 
up the head-waters of the Swat river, which was to carry 
me as far as the present border of the Badshah's dominions. 
Beyond it lie Kalam and other alpine tracts of the Swat 
Kohistan, which, though still independent, may soon become 
a bone of contention between him and the rulers of Dir and 
Chitral. Of course, I had promised to take care not to 
prejudice future developments by going beyond the terri- 
tory actually held by my host. The people of the Swat 
Kohistan are of the same Dard race that inhabited the 
whole of Swat down to the Pathan invasion, and I knew that 
it would be useful to collect specimens of the two Dardic 
tongues still spoken up there, together with anthropological 
data. Another reason that made me eager to push up the river 
as far as I could without giving rise to complications of a 
'political' nature was that the country ahead had never been 
surveyed. So there was plenty of work awaiting TSrabaz 
Khan, my hardy Afridi Surveyor, who was proving an un- 
tiring climber and a very skilful plane-tabler. I had reason, 
therefore, to hope that the Survey, too, would be pleased 
with the results of this northward extension of our tour. 

An easy march past numerous fair-sized villages up the 
amply irrigated main valley brought us to Khwaja-khel, a 
large place where a considerable stream from the watershed 
range towards the Indus debouches. Heavy rain started 
in the evening and continuing practically without a break 
forced us to halt there the next day. This delay was not 
altogether unwelcome to myself; for what with the long 
days spent on our hunt for ruins and 'idols', the late hours 
devoted to detailed diary records, work on photographic 
negatives, &c., it had been difficult to spare enough time 

G2 



84 ON THE WAY TO THE SWAT KOHISTAN CH. xi 

for other duties or for rest either. But I felt sorry for 
those who were looking after us under such weather con- 
ditions. With the exception of a few watchful sentries I 
found them all huddled up sound asleep in their tents, even 
well past midday. 

The- heavy downpour that had kept me all day writing 
in my little tent and had surrounded it with pools they 
marked the sunk-in tombs of an old graveyard fortunately 
ceased on April 4th and allowed a fresh move up the valley. 
For the first two marches the journey brought no marked 
change in the scenery. Though the hills on either side in- 
creased in height and approached much closer to the river, 
they left enough room by its banks for rice cultivation on 
wide terraces. Up to Churrai, which we reached on the 
second day, the abundance of water, coupled with adequate 
summer warmth, allows two crops to be raised in the year. 
The first is always wheat, oats, rape-seed, or a sort of 
lucerne, not unlike the familiar bidar of Turkestan. So the 
eye could feast on delightful stretches of fresh green, and 
spring flowers abounded along the low hedges bordering 
the fields. 

The cultivated belt of alluvium is much narrower on the 
right or western bank, and it was this bank that I followed 
in order to visit the ancient remains of which reports had 
reached me. Each of these visits meant a stiff scramble up 
the hill-side, where thorny scrub invariably covered all 
ground that is not converted into small terraced fields. The 
ruins of Stiipas and monasteries which I here traced were 
modest in size, yet very welcome. They definitely dis- 
proved the prevailing belief that Buddhist remains do not 
extend beyond Charbagh near the big bend of the river. 
At the same time they afforded me reassuring evidence of 
the genuine endeavour of my protectors to secure whatever 
local information could be got about 'old things 1 . 



44 CROWD 01 SI L'DKXTS AXD OTHERS AT GARAI MADRASAH, CHAKtSAR. 

Soi' l)flo\\, page i6r 




CH. xi BUDDHIST REMAINS ABOVE SWAT RIVER 85 

Without their constant help, indeed, it would have been 
quite impossible to trace such ruins, so little was there to 
distinguish the crumbling walls or mounds at a distance 
from the rocks or scrub of the hill-sides. It was still too 
early in the year for fresh grass to appear on the slopes and 
set off the reddish-brown remnants of walls and the heaps 
of potsherds usually found near them. Certain types of 
incised decorative motifs, simple but of good taste, are 
always frequent among the ceramic debris of Buddhist 
times, and approximate dating has thus become easy. 
Small money rewards for ornamented pieces soon turned 
my agile escort into eager hunters for such 'rubbish'. 

At Sambat the tossing river had still to be crossed on a 
skin raft. But below Paitai, where we had to regain the 
left bank for the night's camp, the river-bed had narrowed 
sufficiently to be spanned between high rocky banks by a 
timber bridge. It was as yet under construction, and 
though planned on sound cantilever principles, the three 
big rafters used for its central portion swayed enough to 
frighten 'Moti', the Sipah-salar's intelligent spaniel. He 
absolutely refused to cross on his own feet and had in the 
end to be carried over. The dog had been bought at the 
Mardan cantonment and his parents had obviously been 
accustomed to live with 'Sahibs'. So 'Moti' and I soon 
became special friends. Of course, I talked to him in 
English, and my Pathans seemed to find his apparent com- 
prehension of that 'Wilayatl' tongue quite natural. 

At Paitai an open alluvial fan, at present uncultivated, 
near the Badshah's newly built fort made a delightful 
camping-place. The faint scent of young thyme was in the 
air and somehow carried me back to far-off 'Dashts' on the 
uppermost Oxus or on the glacis of the K'un-lun. The next 
day brought much satisfaction to the antiquarian. Once 
again I had to carry on my archaeological search on the 



86 ON THE WAY TO THE SWAT KOHISTAN CH. xi 

right bank, while the baggage, including the impedimenta 
of my hosts and protectors, kept to the more open ground 
on the eastern side of the valley. 

Fully twenty-eight years have passed since I received from 
Colonel Deane, then Political Agent on the Malakand, the ink 
impression of an inscribed stone brought to him by his above- 
mentioned native agent, then not yet turned into a forger 
of such things. It showed two big footprints marked with 
the Buddha's emblem, the wheel of universal sovereignty, 
and below them a line of bold Kharoshthl characters. I 
sent the impression on at once to Professor Biihler, the 
greatest of Indian epigraphists and Indologists of his time. 
In my accompanying note I expressed the belief that the 
inscribed stone, said to be situated at Tirat, a village of 
Upper Swat, was probably the same that Fa-hsien and 
Hsiian-tsang describe as showing the miraculous footprints 
of the Buddha. Professor Biihler, in the brief notice he 
promptly published of the inscription, a couple of months 
before he lost his life on the Lake of Constance, showed 
that I was right in this conjecture; for the legend as de- 
ciphered by him distinctly describes those footprints as left 
by the Enlightened One when he rested at that spot. 

But even before I could visit Tirat and the sacred relic 
still surviving near it, my antiquarian heart was to be grati- 
fied by another discovery. Both those old Chinese pilgrims 
mention another sanctified spot some distance south of the 
stone with the footprints. It was a big rock by the river 
bank, on which pious eyes could still see the imprint left 
by the clothes which the Buddha had washed and then 
dried there. My local informants had told me of a 'big 
stone with writing' to be seen on the way to Tirat village 
on the right bank of the river, and naturally my thoughts 
had turned at once to this second notice of my patron saint. 
So my satisfaction was great when on descending from 










X K 

O -" 



H C 
H X 



O , 

P 3 



X ,<* 

G S 



CH. xi ROCK OF BUDDHA'S CLOTHES-WASHING 87 

some ruins of a small Buddhist shrine high up in the Ragast 
Nullah I was taken to the spot indicated. 

There by the roadside rose a large mound marking a 
completely demolished Stupa, and some hundred yards off 
above the deep-cut rocky bed of the river I was shown 
a huge boulder, some thirty feet long and twelve feet high 
(Fig. 48), covered on its fairly smooth southern side 
with numerous graffiti in Sanskrit and also in Arabic char- 
acters. The former were so weathered and effaced that I 
was unable to decipher more than a few auspicious syllables ; 
but these sufficed to show that these scribblings went back 
to early medieval times, when Buddhism in its Hinduized 
form still prevailed in these parts. The Arabic graffiti, 
most of them repeating the Islamic creed or Kalimah, also 
had their interest, as attesting continuity of local worship. 
But of the holy markings left by the Buddha's garments no 
trace remained except the place which they had once 
occupied. Over a space fully eleven feet high and eight 
feet wide the smooth surface had been completely scraped 
off. Was this the handiwork of some fervent Mullahs, 
propagandists of Islam, which is known to have penetrated 
into the Swat Kohistan only some eight or ten generations 
ago, or of relic-hunting devotees of old ? 

Curiously enough, local worship had here derived fresh 
nourishment from a recent event. Between the Stupa 
mound and the road I found a grave mound marked by 
votive offerings of various kinds. It held the remains of an 
unfortunate Afridi who some two or three years before had 
come as a trader of rifles, a very valued commodity in these 
parts, and had been murdered not far off by some men from 
the Kohistan. His grave was appropriately placed here and 
is now receiving due worship as the resting-place of a 
Shahid or martyr. If life is of little account as yet in this 
region, there is anyhow some compensation to be found in 



88 ON THE WAY TO THE SWAT KOHISTAN CH. xi 

the ease with which a violent death converts the sinful 
victim into a martyr. 

A stiff climb over thickly wooded slopes, where I met the 
first cedars intermingled with firs and wild olives, took me 
from the 'clothes-washing 1 site up the picturesque side 
valley that holds the scattered hamlets of Tirat. There on 
a terraced field, facing the largest of them across a deep-cut 
stream, I was duly shown the inscription, engraved on the 
smooth face of a big fragment of rock. Turned sideways 
and half-hidden in the ground, it now serves as part of a 
stone fence by the side of a narrow lane (Fig. 47). The two 
footprints, though engraved only to a slight depth, might 
well seem impressive, being fully nineteen inches in length. 
But how 'the size varied with the religious merit of the 
measurer', as Hsiian-tsang's 'Memoirs' tell us, it would be 
difficult now to guess. 

The shape of the stone left no doubt that it was placed 
originally with its flat surface upwards. It had evidently 
been thrown out when the shrine which held it was de- 
stroyed. Luckily the owners of the adjoining fields, though 
Mullahs and hence men of religion, had seen nothing in the 
stone to offend their orthodox feelings or to mislead their 
flock. Otherwise the stone might well have been defaced 
or conveniently reburied when it was accidentally brought 
to light again many years ago. The impression sent to 
Biihler in 1897 was a good one. So I could content myself 
with using the remaining daylight to take photographs and 
rapidly survey two ruined mounds close by. One of them, 
some forty feet across its flat top and of unusual shape, 
doubtless hides the remains of the shrine that our saintly 
Chinese guide mentions as having been erected over the 
miraculous footprints. 




$. BOULDER MARKING Sl'OT OF 'BUDDHA'S CLOTHES-WASHING', 
ON RIGHT BANK OF SWAT RIVER 




CHAPTER XII 

THE ENTRY INTO TORWlL 

As the night fell we made our way from Tirat down to the 
river, some eight hundred feet below, and crossing once 
again to the left bank reached the large village of Churrai. 
It is the gate to the Swat Kohistan or Torwal, the name given 
to the portion of the Swat valley absorbed four years before 
in the Badshah's dominion. Churrai, with its closely packed 
dwellings and lanes scarcely wide enough for two ponies to 
pass, is quite a busy place of local trade. From there come 
all the closely woven and gaily but tastefully coloured 
woollen blankets that India knows as 'SwatI rugs'. They 
are made by the womenfolk in the side valley of Chihil-dara, 
which descends to Churrai from the high snowy peaks to- 
wards Kana and Duber on the east, and to some extent 
also in other smaller valleys of Torwal. 

That this local industry of TSrwal is as ancient as the 
Dard race that retains its hold there is proved by a passage 
which M. Sylvain Ldvi, the eminent French Indologist, has 
quoted in his comments on a curious Buddhist Sanskrit text 
published by him under the title of La geographic des Yakshas. 
It contains lists of demons whom pious legend associated as 
a kind of genii loci with particular localities in different parts 
of India and especially in its Frontier region. In that 
passage which by its Chinese translations is shown to belong 
to the early centuries of our era, we already find the rugs of 
U4<jliyana or Swat mentioned by the same term of kambala 
that is now applied in Indian vernaculars to these textile pro- 
ducts of Torwal. Indian literature, so strikingly deficient, 
alas, in records of real history or geography, can scarcely 
contain any earlier testimony to the antiquity of a still 
flourishing local industry. 

Unfortunately, though the ancient skill in weaving and 



90 THE ENTRY INTO TORWAL CH. xii 

the use of traditional patterns still survive, the introduction 
of aniline dyes has here, as elsewhere in the East, brought 
about a sad and rapid decline in the harmonious blending 
of colours. Rugs produced with the fine old vegetable 
dyes, such as were still obtainable at Peshawar some thirty 
years ago, could now no longer be found for me even in the 
remote tract where this manufacture has had its home for 
so many centuries. 

At Churrai the chance of collecting my first specimens 
of Torwali, a Dard language so far practically unrecorded, 
and of taking anthropological measurements among its 
speakers, detained me for a day (Fig. 49). This occupation 
was pleasantly varied by a walk through the narrow lanes, 
where many doors and shop recesses showed excellent wood- 
carving (Fig. 50), and along the lively stream which comes 
down from the side valley of Chihil-dara. Upper Swat 
abounds in life-giving water, and now that the melting 
snows were filling all these mountain torrents it was a con- 
stant feast for the eyes to watch their tossing white cascades. 

The sound of the rushing water and of the rapids of the 
young Swat river was loud enough. But it was by no 
means sufficient to drown the local noises at night-time. 
They came not only from the wide-awake cocks and babies 
of the village in the immediate vicinity of which we had 
been forced to camp for want of other space. A considerable 
proportion was contributed by my protective force, some 
forty men in all, as a result of the Ramazan fast. What with 
the evening meal put off until nightfall and the necessity 
of taking the only other meal permitted by sacred law long 
before the first dawn, the men seemed to think it scarcely 
worth while to take any sound sleep between. Preparations for 
cooking the early meal usually began about midnight, and 
there were always little groups chattering while they eagerly 
watched the process. How, with so little regular sleep, the 




50. WOOD-CARVINGS ON DOOR OF HOUSE, CHURRAI. 




5i. BRIDGE OVER SWAT RIVER AT AlN, BELOW BRANIAL. 



CH. xii TORWALl-SPEAKING HILLMEN 91 

men were able to keep up a brisk walk on the day's march 
was a marvel. There is plenty of lively resilience in Pathan 
physique and mentality, and the capacity for taking sleep 
in snatches at any time probably proves useful not only to 
great men like Napoleon but to humbler mortals also. 

From Churrai onwards the valley rapidly contracted and 
the scenery soon began to recall that of the Jhelam valley 
below Kashmir (Fig. 51). Glorious weather favoured the 
first march, which brought us past picturesque hamlets high 
up on wooded spurs to Branial, the chief place of the 
Torwall-speaking hillmen (Fig. 52). But there our luck 
changed. Two days of rain, at times quite torrential, kept 
us at Branial in conditions not altogether enjoyable. The 
narrow terrace outside the newly built fort where our camp 
had of necessity been placed was soon thoroughly soaked, and 
for even a few steps outside my little tent I needed the stout 
mountain boots that I had brought from Oxford. 

Fortunately I was able to employ this long halt to good 
purpose, overtaking the formidable arrears of correspon- 
dence with which my constant moves were threatening to 
submerge me, and recording specimens of Torwall stories 
for publication by my friend Sir George Grierson, in 
connexion with his monumental Linguistic Survey of 
India. I am no expert in Dard philology and not endowed 
with a very keen ear. So it was doubly fortunate that the 
Branial reciter whom Raja Shah *Alam secured for me was, 
like most of his fellow Torwalis, a person rather slow of 
tongue and brain, and the composition of his tales accord- 
ingly simple. The constant repetition of phrases but slightly 
varied served me well in a quasi- Ollendorfian fashion, and 
in the end the word-for-word translation of the record taken 
did not prove so difficult a task as I had at first feared it 
would be on account of my imperfect command of Pashtu. 
It was through this medium that all our conversation had 



92 THE ENTRY INTO TORWAL CH. xn 

to be carried on, and neither Mukadir Shah, the Torwali, 
nor Shah c Alam, the Khushwakt from Yasin and Tangir, 
nor myself had a thorough mastery of Pashtu. 

But Shah c Alam combines with other excellent qualities 
a real linguistic ability, and Chitrali, Shina, Kohistani, all 
of which h6 fluently speaks in addition to Persian, Pashtu, 
and a little Hindustani, are, as Dard languages, allied to 
Torwali. The latter, like the other dialects of old Dard 
stock spoken by independent tribes in the Indus gorges, is 
bound to be swept away in course of time by the ever 
advancing tide of Pathan speech. So I felt some gratifica- 
tion at this opportunity of adding to the extremely scanty 
knowledge so far available of this linguistic 'relict'. The 
total population of Torwal can scarcely exceed six thousand 
souls. Significantly enough it includes quite a number of 
immigrant families speaking Chitrali or Pashtu. And in 
addition to all this mixture, there are numerous Gujars, 
familiar to me from residence in Kashmir, grazing their 
cattle high up in the side valleys and cultivating on lease 
such little plots as Torwalls can spare them. 

But my halt at Branial enabled me to make interesting 
observations also in another direction. When I visited the 
village on the day of my arrival, I was greatly struck by 
the amount of fine wood-carving, old and new, to be seen 
in this quaint rabbit-warren of houses (Fig. 55). The 
lanes between them are incredibly crooked and narrow. 
But a curious method of 'town planning' secures unex- 
pected open spaces owing to the fact that the flat roofs of 
certain groups of these houses, most of them of two stories, 
are combined into little piazzas. These roofs are made 
accessible to fellow citizens or strangers by primitive stairs 
or ladders. The houses are built mainly of timber with 
wattled walls, and recalled those of the early centuries after 
Christ which I had excavated from the sands of the Tak- 




H 








53. WOODEN PILLARS WITH CARVINGS IN LOGGIA OF PRINCIPAL 
MOSQUE, BRANIAL. 



CH. xii WOOD-CARVING OF BRANIAL HOUSES 93 

lamakan at the Niya Site and elsewhere. My general im- 
pression was that the methods of building and living in this 
mountain tract, difficult of access and so little exposed to 
outside influences, cannot have changed greatly since the 
times when Lower Swat with its fertile lands enjoyed its 
flourishing civilization of Buddhist times, or when, even 
earlier, Bactrian-Greek chiefs, bringing Hellenistic tra- 
ditions in speech and material culture, held sway there. 

This earlier phase in the history of the North- West 
Frontier was vividly called up before my eyes by the re- 
markable variety of decorative motifs of purely Graeco- 
Buddhist art displayed in the abundant wood-carving 
on the doors, pillared verandahs, &c., of Branial houses 
(Figs. 53, 54). This is not the place to discuss details. It 
must suffice to state briefly that many of the most frequent 
designs, floral scrolls (often including the acanthus), bands 
of ornamental diapers, &c., with which I was so familiar 
in the Gandhara relievos, as well in wood-carvings at 
ancient sand-buried sites in the Tarim basin, have survived 
practically intact in this far-off corner of the mountains. 
Distinct traces of the influence once exercised by Graeco- 
Buddhist art in the Indus region had previously forced 
themselves upon my attention in Chitral, Darel, and else- 
where in the Hindukush region. But nowhere was evidence 
of it so striking as in this surviving craft of the TSrwal 
wood-carvers. I did my best to secure specimens of it for 
our New Delhi Museum of the future in the shape of old 
carved brackets and panels thrown away as useless lumber, 
and also had rough drawings made by a local craftsman. 



CHAPTER XIII 

TO THE HEADWATERS OF THE SWAT RIVER 

ON the nth April I was able to set out again up the valley 
in order to reach the extreme northern limit of the Bad- 
shah's dominion. From Branial onwards the hardy Swat 
mules had to be replaced by human transport (Fig. 64). 
The ruler had lost no time after his annexation of Torwal 
in getting a road made practicable for ponies and mules. 
But it would not have been safe to use it for laden transport 
after the damage caused by the recent heavy rain and, in 
some places, by avalanches. The two marches that carried 
me to Peshmal, the Badshah's last village on the Swat 
river, led through scenery of quite an alpine character. The 
route led up and down along steep spurs and across pic- 
turesque side valleys flanked by mountains still snow- 
covered (Fig. 56). No archaeological remains were to be 
looked for in such surroundings. But it was delightful to 
feel that I was moving between mountains on which no 
European eye had hitherto rested, at least not since classical 
times. To the east, where the high watershed range divides 
the Swat river drainage from the Indus, snowy crests and 
peaks, like walls of crystal as the old Chinese pilgrim 
Sung Yiin put it gleamed at the head of each narrow side 
valley (Fig. 57). 

All the higher slopes were clothed with magnificent coni- 
fer forest up to 9,000-10,000 feet, and from Ramet onwards, 
about half-way up on the first march, deodars and then 
firs and pines of magnificent stature became more and more 
frequent, even along the 'road*. There was, unfortunately, 
much evidence of the destructive methods of work of the 
Kaka-khel sept of Nowshera, those saintly timber-contrac- 
tors who for years had had a free hand in their dealings 
with the Kohistams of Swat and elsewhere. Some of the 




a < 
m 
o w 




CH. xni TO CHODGRAM AND PESHMAL 95 

more accessible spurs had been badly denuded, while in 
places huge trees had been felled and left unsawn owing to 
some squabble with the Torwali wood-cutters. All this 
destructive exploitation has happily been stopped since the 
ruler of Swat established his hold on the Kohistan. He 
realizes the value of its timber resources and is setting about 
their utilization in a more intelligent fashion. 

We halted for the first night at Chodgram, under the 
protection of a small timber-built fort (Fig. 58) quite 
antique in appearance, though new. Here a glorious view 
opened over the Mankial valley and the bold ice-crowned 
peaks that close it to the east. They were the same great 
peaks that I had first sighted far away from the Malakand, 
rising to heights not far from 19,000 feet. Chodgram itself 
lies on a spur high above the confined gorges of the river, 
and this made the vista wonderfully wide and impressive. 
I only wish my photographs could have done full justice 
to it (Fig. 59). 

On the next day's march to Peshmal we had to keep for 
the most part low down by the river. But there was plenty 
of true alpine scenery on which to feast one's eyes, to say 
nothing of the lovely colours of the river water, which 
showed every shade of green, and of charming glimpses up 
the cascading streams that come down to feed it. No more 
villages were to be seen here, only scattered homesteads 
clinging to the green slopes, where water could be brought 
to irrigate small terraced fields. It had been pleasant to 
note, as we moved higher and higher up the valley, how 
the number of fruit-trees increased. We had been travelling 
in the wake of spring, and at Churrai the apricots and apple- 
trees were already in full bloom. But up here, at elevations 
of 6,000 to 7,000 feet, only little flowers in sheltered spots 
announced the coming of spring, and the big walnut-trees 
still stood as bare as they did a month ago in Kashmir. 



96 THE HEADWATERS OF THE SWAT RIVER CH. xin 

From Peshmal, a rude Gujar hamlet, I enjoyed a grand 
sight of the big boldly serrated snowy spur thrown out 
northward by the high massif above Mankial (Fig. 60). 
Glaciers of some size descend from this spur. But what 
delighted me even more was the vista from our high-lying 
camping-place towards the north. There a grand wall of 
snowy mountains rose before me, and I knew that it marked 
the approach to that distant Hindukush region which holds 
Mastiij and Yasln, full for me of happy recollections of my 
travels in 1906 and 1913. 

Next morning, with a temperature only a little above 
freezing-point, I made an early start with the Sipah-salar 
and his host. We were to push ahead to the ridge of Korun- 
duke, which marks the boundary between Torwal and the 
still independent tribal community of GarwI-speaking hill- 
men who hold the headwaters of the Swat river. I had 
been promised from there a full view of Kalam, the big 
village situated at the meeting-place of the two main 
branches of the Swat river, which come from Ushu and 
Utrot. The hope was not disappointed. When the ridge 
was reached after a rapid climb of three miles, a glorious 
panorama lay before me (Fig. 61). 

The reports that credited Kalam with a big expanse of open 
arable land were true. Nestling below an amphitheatre of 
mountains all still carrying snow lies a fertile little plain, 
which seems to mark the basin of an ancient lake scooped 
out probably by glacier action. To eyes that had grown 
accustomed to deeply eroded gorges this plain looked quite 
imposing. Above it, only a little over two miles from our 
ridge, stretched the long lines of houses, built in tiers, that 
form the chief place of Kalam, 

It looked quite an inviting place from that distance, with 
dense forest coming down to it over a gentle glacis of the 
sheltering range. Such might some prosperous little town 




CHODGRAM VILLAGE SEEN FROM SOUTH : SNOWY RANGE 
ABOVE KALAM IN DISTANCE. 




59. SNOWY PEAKS ABOVE JABA VALLEY SEEN FROM ABOVE 

CHODGRAM. 
On extreme left is seen pass of Chonkur-kandao leading to Duber valley. 



CH. xiii IN VIEW OF KALAM 97 

appear at the foot of the Swiss or Bavarian Alps. It was 
for me a strangely elevating sensation that though further 
progress was barred by 'political considerations', I had 
sighted land that seemed by comparison next door to 
Chitral, Yasin, and Tanglr. Once again it had fallen to 
my lot to approach by a new route the huge barrier of the 
Hindukush, that true divide between India and Central 
Asia. 

I had brought with me Torabaz Khan, my indefatigable 
Surveyor, in order to make sure that due justice should be 
done to the geographical interest of the still forbidden 
ground before us. The big peak of Mankial, 18,750 feet in 
height and locally known as Koshujan, proved to be the only 
trigonometrically determined point within the fine panoramic 
view to the south (Fig. 62). But together with lower peaks 
observed on our way up it sufficed to fix our position with 
a fair degree of accuracy. The three hours we spent there on 
survey and photographic work will long be remembered by 
me with joy. Thin layers of snow still covered the ridge on 
its northern face. But wherever the sun's warm rays reached 
sheltered spots among the rocks pale blue violets were to be 
found in profusion. They had little or no scent, these dear 
messengers of primavera. Yet I greeted them with quite as 
great delight as those first wild cyclamens I had picked a 
year before on the wooded crest above the lake of Albano. 
Once again our globe seemed to me to become strangely 
small. 

Reluctantly I hurried back the same day to ChSdgram 
to bid au revoir to those glittering peaks above Mankial. 
Thence a day later we returned to Branial. It had to be 
reached in time for the due celebration of the 'Akhtar' or 
Id, which marks the close of the Ramazan. A day's halt in 
honour of the great feast allowed me to give a good treat 
to my Swat friends and the hardy escort (Fig. 63). It 



98 THE HEADWATERS OF THE SWAT RIVER CH. xin 

was a pleasure to see how quickly the big heaps of pilaw 
disappeared among three score of mouths which for a 
month had not touched food or drink in daylight. I was 
also able to use this halt to secure further anthropological 
measurements of Torwalis (Fig. 52). 

On April i6th the return journey down Torwal was re- 
sumed. It seemed cruel to withdraw my protecting host 
from the flesh-pots of hospitable Branial ; for the Id feasting 
had continued long into the night, and seeing that it followed 
such long privations, some relaxation of spirits was in- 
evitable. All the same a double march was accomplished 
to Paitai. A short half-way halt at Churrai allowed me to 
acquire some fair specimens of those gay-coloured 'SwatI 
rugs' to which I have already referred. Long before I 
could dispatch them as presents to distant friends, I had 
myself occasion to feel grateful for the warmth of these 
heavy woollen coverings. 




[ at plane-table in right bottom corner. 



CHAPTER XIV 

ACROSS THE SWAT-INDUS WATERSHED 

I WAS now anxious to get to work east of the Swat- Indus 
watershed, and therefore decided to make my way across 
it by the nearest pass that might be expected to be sufficiently 
clear of snow for laden transport. A second march led 
down the left bank of the river and allowed me to see more 
rock-carved Buddhist images (Fig. 66). Towards the end 
of it we turned up the big valley that debouches at Khwaja- 
khel and reached the village of Shalpin below the Karorai 
pass. Light showers had accompanied us during most of 
the morning and again later caused a hasty break-up of the 
roadside halt during which my Swati protectors enjoyed 
their tea (Fig. 65). But after a warm and sunny after- 
noon I was little prepared for the heavy downpour that 
followed a thunderstorm late in the evening. Accompanied 
by violent wind, it continued all through the night. My 
fear that the tent would collapse through the yielding of the 
wooden pegs in the sodden soil was fortunately not realized. 
But my dismay was great when in the morning I found 
that a little stream of liquid mud had made its way under- 
neath my camp-bed and had passed by the cameras. How 
it failed to penetrate them was a wonder to be accounted 
for only by the solid English leather of which their cases had 
been made some twenty years ago. The path above the 
camp had been turned into a watercourse by the torrents 
of rain, and when the side of it gave way had spread its 
muddy stream over the terrace below. 

The next day, April i8th, was wretchedly wet, and all I 
could do in the midst of rain and deep mud was to attend 
to heavy arrears of writing. On the morning of April igth 
the sun at last struggled through the mist and clouds that 
shrouded the mountains. We were only too glad to leave 

H 2 



ioo ACROSS THE SWAT-INDUS WATERSHED CH. xiv 

our sodden terrace as soon as the tents were half-dried. 
Once again I admired Jthe cheerful goodwill of our escort. 
The Badshah's 'men-at-arms' clothe themselves as best 
they can. This means that few of them have more than a 
single suit of thin cotton garments, with perhaps a large 
sheet 6f the same material to sleep under. The solid leather 
bandoliers or more elaborate Sam Browne belts, neatly 
packed with cartridges, that they all display do not help to 
give protection from cold and damp. Fortunately Pathan 
villagers observe the hospitable law of their race and pro- 
vide strung cots and cotton 'Razais' so far as they can. 
Failing these their Gujar mttayers, characteristically 
called Fakirs, i.e. beggars, together with all other folk 
cultivating but not owning land, are called on to contribute 
from their humble belongings. But on such nights I felt 
doubly uneasy for the sentries who kept constant guard 
over my tent and person. The only thing I could do to 
alleviate their lot was to lend two waterproof sheets which 
could be spared from my own men's equipment. 

The well-graded bridle-path, one of the network of 
'strategic roads' with which the Badshah's wise forethought 
has spanned his newly consolidated kingdom, made the 
ascent to the Karorai pass easy for the first seven miles. 
Then we got into soft fresh snow and, when the top was 
reached at an elevation of about 6,400 feet, the ridge clothed 
with fine firs and pines presented as wintry an aspect as I 
could wish to see after a rainless 'cold weather' in Delhi. A 1 
thick white mist was clinging to both sides of the ridge, and 
after trudging with Torabaz Khan, the Surveyor, for about 
half a mile along its fairly level crest, we had to abandon 
the hope of making a 'fixing' with the plane-table. We 
could get only passing glimpses through rifts in the mist 
into the maze of deep-cut gorges draining eastwards into 
the Ghorband river and thus into the Indus. 




(>3- FEAST AFTHR RAMAZAX AT BRANT AT*. 

Seated in ceritie. Sipah-salar Ahmad 'All, 'Abdul Latif Khan, and Raja Shah 'Alain. 





05. ROADSIDE HALT FOR TEA OK SIPAH-S \I-AR AND 1-SCORT, AKO\ E 
KHWAJA-KHLL. 




CH. xiv ACROSS THE KARORAI PASS 101 

The setting-up and reading of the mercurial mountain 
barometer on the pass was a trying operation, for an icy 
wind was blowing at a temperature just about freezing- 
point. When this, in spite of benumbed fingers, had been 
safely accomplished I was glad to hurry down to where, 
half a mile or so beyond, the men of the escort were warming 
themselves round a fire at the foot of an old pine-tree. Here 
the two little aluminium tins containing my ' tiffin", warmed 
over the embers, provided welcome comfort. 

A mile or so farther the snow was left behind, and the 
mist lifting from the slopes below revealed a landscape that 
looked strangely bare and stern under a grey sky after the 
green expanse of the open Swat valley and the alpine 
charms of Torwal. Only a thin growth of firs and cedars 
could be seen on the upper portion of the slopes, while such 
villages and hamlets as might be found in the deep gorges 
of Ghorband were hidden from view. Our goal was the 
main village in the side valley of Lilaunai, which joins that 
of Ghorband from the north near the sharp bend that the 
latter makes eastwards. At last it came suddenly into view 
as the path, by a narrow gap, crossed the shaly spur that 
we were following. 

A steep descent to the confined bottom of the valley only 
helped to confirm that first sombre impression. Owing to 
the elevation, about 5,300 feet above sea-level, and still 
more, perhaps, owing to the high range which overlooks 
Lilaunai from the south and reduces the sunshine, vegeta- 
tion down below still seemed very backward. The big 
plane-trees and elms that guard a Ziarat and burial-ground 
on the right bank of the considerable stream had not yet 
put on leaves. Otherwise it seemed the finest grove of 
leafy trees that we had so far seen. Swat and adjacent 
tracts are singularly devoid of arbours or orchards such as 
the living might enjoy, a fact attributable, as we have seen, to 



102 ACROSS THE SWAT-INDUS WATERSHED CH. xiv 

the Pathan system of wesh or periodical redistribution of 
lands among the families belonging to the same tribal 
sub-division. 

But the dead fare better in this respect. Around and 
over their graves thickets of ilex, wild olive, and other trees, 
accordiftg to the nature and elevation of the ground, are 
allowed to grow up. Time turns righteous and wicked 
alike into 'Shahids' or holy martyrs, and the sanctity thus 
bestowed upon burial-places protects the trees from the 
villagers' otherwise reckless axe. In fact, these burial- 
groves become sanctuaries for whatever timber, hay, &c., 
may be deposited there by individual villagers for future use. 

What was to be seen at Lilaunai, after we had reached it 
by a bridge newly built by the Badshah's order across the 
tossing stream, did not help to relieve its first rather gloomy 
aspect. As the Sipah-salar, who knows all his master's con- 
quests like his pocket and has done much to secure them, 
explained to me, the village had suffered to an unusual 
extent from internal feuds, until the Badshah some four 
years before swept into his net all these valleys between the 
Indus and the Swat watershed. Ruined homesteads, roof- 
less houses, cut fruit-trees met the eye in more than one 
place before we reached the Badshah's newly erected fort, 
quite medieval in structure, which now enforces peace (Fig. 
67). The Papini Saiyids, who hold Lilaunai, claim saintly 
descent; but this does not, of course, prevent them from 
dearly cherishing their enmities and from being much given 
to violence otherwise. So the Badshah has wisely put a 
capable man here, chosen from among his old supporters, 
to hold combined military and civil control ; a true ' Vogt' 
as they called them in Switzerland in old times. He in turn 
has taken care to get a comparatively strong fort built, of 
course, by the Saiyids themselves. 

The good quarters provided on its upper floor proved a 




67. NEWLY BUILT FORT AT LILAUXAI. 
U.iggagc bring loaded 111 foreground 




CH. xiv WEATHER-BOUND AT LILAUNAI 103 

boon to my Swatl friends and their myrmidons; for the 
night brought more rain, and this, with short interruptions, 
continued all through the following day. My dripping 
tent, closely surrounded by guard and kitchen tents, timber- 
stacks, odorous refuse-heaps among boulders, &c., was 
not exactly an ideal place for writing. But I was able to 
do a good deal of it, the roar of the torrent just below 
drowning all other noises. The Shilkai pass, by which I 
had hoped to reach the head of the Kana valley to the 
north-east, had been reported still blocked by heavy snow. 
But the Sipah-salar, keen soldier that he is, with a good 
eye for the map that he carries in his head, realized the 
time that would be wasted and the opportunities for survey 
that would be lost if I had to change my proposed route for 
a detour down the Ghorband valley. So messengers were 
dispatched to the Kana Khans for a contingent to clear a 
way through the snow from their side while the com- 
mandant of Lilaunai sent up gangs of Gujars and mis- 
cellaneous 'Fakirs', that miser a contribuens plebs, from 
his own. 

A tramp down the rapidly narrowing gorge of the Li- 
launai stream over dripping cliffs and soaked cultivation 
terraces, in search of a reported 'written rock', proved fruit- 
less; for the markings borne by the big boulder were 
natural. But the climb that I undertook next day with the 
Surveyor in order to resume map-work proved more fruit- 
ful. By a three hours' steady climb a good survey station 
was reached on the high spur of Jabo-sar dividing the lower 
portions of the Lilaunai and Kana valleys. It provided 
a good distant view towards three or four high snowy peaks 
to the north that had previously been triangulated on one 
of the Black Mountain expeditions across the Indus; and 
we were fortunately able to complete a round of theodolite 
bearings to unsurveyed heights south and east by the early 



104 ACROSS THE SWAT-INDUS WATERSHED CH. xiv 

afternoon, when the steadily gathering clouds blotted out 
once more all distant views. 

I thought it an encouraging omen that the view to the 
south had allowed me to catch a short glimpse of the long 
thickly wooded range that stretches from the head of the 
Ghorbahd valley due east to where the Indus makes a big 
bend round its foot above Thakot. It was to that end of 
the range that my lamented friend Colonel Wauhope, late 
of the Survey of India, on the strength of the impressions 
he had gained at Thakot in 1892 on the last Black Mountain 
expedition, had directed my attention when years before we 
discussed at Dehra Dun the problem of Aornos. The in- 
quiries subsequently made on my behalf by Colonel James, 
then Political Agent on the Malakand, pointed to the ridge 
known as Pir-sar, an eastern outlier of that range, as a 
likely position for the famous rock fastness that had so long 
been searched for. But not much reliance could be placed 
on the reports of native informants, naturally eager to 
supply the 'Mulki Sahib' with a suitable answer to his 
queries, and only by an actual survey could the question be 
cleared up. 



CHAPTER XV 

OVER THE SHILKAI PASS AND DOWN K&NA 

ON my return to the Lilaunai fort I was welcomed by the 
report that the pass to the head of Kana had been made 
practicable for load-carrying men on the morrow, provided 
no fresh snow fell. We all prayed that this condition might 
be fulfilled, and a perfectly clear morning on April 22nd 
brought the anxiously awaited chance. The baggage had 
to be divided into as small loads as practicable and distri- 
buted among the big crowd of men collected for the crossing. 
So it was nearly 10 a.m. before a start could be made. 
Under a brilliantly clear sky the narrow side valley up 
which the route led to the north-east looked quite bright by 
contrast with the gloomy aspect of the preceding three 
days. A couple of mosques, adorned with effective if rather 
crude wood-carvings on columns and brackets, had escaped 
the general ruin that prolonged local feuds had brought 
upon the valley. Spring had now at last arrived, and the 
lower slopes on either side, bared long ago of the conifer 
forest that covered all the rest of the mountains above us, 
had quickly responded to its call, gladdening the eyes with 
their first fresh green. 

The snow-covered pass came into full view after five 
miles of easy progress, and three more brought us to its 
foot on a small alpine meadow. The fresh snow was just 
melting here; farther on it made the ascent to the pass over 
the grassy slopes very troublesome. The bridle-path on the 
opposite or southern side of the gully was completely 
obliterated under heavy masses of snow, and a new 
track had therefore to be followed. Passing clusters of 
load-carrying men who took rest wherever the ground 
under trees gave a better foothold, we at last, about 2 p.m., 
reached the crest by a narrow track. Men sent ahead 



io6 THE SHILKAI PASS AND KANA CH. xv 

had trodden it in the snow, which lay here four to five feet 
deep. 

It was strange to be met towards the close of the climb 
by sounds of weird music, produced by drum, fife, and 
bagpipe. It was the little troop of the Khan of Bilkanai 
who had Srought up the three performers, along with a large 
posse of carriers to relieve the Lilaunai men struggling up 
through the snow. The effect was distinctly encouraging, 
and long before I had finished my photographic work and 
modest 'tiffin 1 , all the loads had changed hands and were 
rapidly being passed down the steep slopes towards Kana. 
So eager were the Lilaunai men to get clear of their burdens 
and of the bitterly cold wind blowing over the crest that the 
headmen would scarcely wait for the special reward that I 
wished to have distributed among the carriers. Aneroid and 
hypsometer indicated a height of approximately 9,400 feet. 

From a little hillock above the track, which being clear 
of trees was covered by heavy snowdrift, a fine but limited 
view was obtained to the south. It extended over most of 
the Indus-Swat watershed, with Mount Ham, that great 
landmark of Swat and Buner, far away in the distance. 
But a much grander panorama opened to the north and 
north-east; there a continuous wall of snowy heights, with 
peaks from 14,000 to 15,000 feet, encircles the head of the 
Kana valley, which spreads before the eyes like a huge 
amphitheatre (Fig. 69). Beyond that glittering white wall 
lay the unexplored high valley of Duber, belonging to 
the Indus Kohistan. The steep pass that gives access to it 
could be seen by the side of a prominent peak. The narrow 
avalanche-swept gully below it, as seen through my glasses, 
seemed impracticable at this season. Yet on our descent to 
Kana we met a small band of sturdy Duber men who, with 
loads of salt and other merchandise from Swat, were on their 
way to attempt it in spite of the abnormally late snowfall. 




(><). VIEW FROM SHILKAI P\SS TOWARDS SNOWY R\\<,K Al HLAD OF KANA 

\ \LLLY 




CH. xv CROSSING OF SHILKAI PASS 107 

For over two miles down from the crest progress meant 
a succession of jumps from one deep hole trodden in the 
snow into the next or else immersion in soft snow. Then 
came a mile of wading in slush or small streams draining 
the rapidly melting fresh snow. At last dry ground was 
reached near the Gujar huts of Charrai, and the scramble 
down over trackless stony slopes, tiresome as it was, seemed 
a welcome change. At an elevation of about 7,000 feet we 
gained the bridle-path that comes down from the proper 
pass. It impressed me as quite a respectable piece of recent 
local engineering, as it winds along the cliffs that line the 
bottom of a narrow deep-cut valley for close on five miles. 

On the steep slopes above no forest was to be seen, but 
cultivation terraces rose in tiers to a height of a thousand 
feet or so above the rock-cut channel of the stream. Here 
and there rib-like ridges running down from the big spurs 
on either side of the valley bore clusters of storied dwell- 
ings. No vegetation as yet clothed the slopes between 
them. In the evening light the whole valley in its monoto- 
nous hues of dark brown and deep red looked distinctly 
gloomy. The sharp outlines of the dwellings clinging to 
step-like terraces high above us, instead of relieving the 
gloom, only helped to produce the forbidding effect of a 
Dor landscape. 

At some places little crowds of villagers who had descen- 
ded from their heights were watching our progress below 
the cliffs. Evidently this visit of ours was a great and 
unwonted event for this far-off valley. Perhaps it also 
afforded some little compensation for the hard work which, 
evidently in view of it, had been put into repairing the 
Badshah's road at particularly difficult points below 
rock faces. At one place we came upon a small 
well-set-up cheerful people from Duber in the inc 
Indus Kohistan. They were about to return to 

J 




io8 THE SHILKAI PASS AND KANA CH. xv 

with big loads of salt and cotton fabrics, which they had 
carried up from the shops of Charbagh. To see the respect 
and delight with which my Khushwakt friend was greeted 
by these sturdy hill-men, next-door neighbours to the great 
valley of Kandia where Shah c Alam with the family of his 
murdered uncle and chief had first sought refuge after the 
upheaval in Tangir, was a touching proof of the attach- 
ment still enjoyed by him as an exile from his kingdom in spe. 

It was with a feeling of relief that below the hamlet of 
Larai I turned at last into the main valley of Kana. Here 
at its head, below that great amphitheatre of snow-covered 
mountains, it proved unexpectedly wide and open. Still 
less was I prepared for the grand fashion in which I found 
our camp laid out at Bilkanai. Dost Muhammad Khan, 
the son of Amir Khan, who holds upper Kana by a kind of 
feudal tenure, is acting now as 'Tahsildar' or bailiff for the 
Badshah. In order to show due consideration for his 
master's chief commander and his host, he had enclosed the 
terraced field where our tents were pitched with lines of 
young pine-trees brought down from the mountains. They 
were hung with strings of gay paper flags in true Indian 
fashion. 

With paths neatly laid out between the tents, the whole 
scene presented a striking contrast to the grim old fortified 
residence of the Khan, under the protection of which the 
camp was pitched (Fig. 71). The construction of the little 
stronghold was peculiar. High solid square walls, without 
any openings except loopholes, contained quarters built 
round a high square keep. Nor were there openings on the 
roof of the quarters, and as from regard for the Khan's 
household I could not proceed beyond the dark entrance- 
hall which serves both as stable and followers' quarters, 
the arrangement for lighting and airing this safe baronial 
retreat remained rather puzzling. 



CH. xv WELCOME IN UPPER KANA 109 

I spent a very pleasant day at Bilkanai. While the 
Surveyor proceeded to a high isolated hill up the valley, I 
examined the remains of ancient dwellings that spread some 
distance up the spur known on account of them as Kandaro- 
sar, and then climbed to its summit at a height of about 
8,800 feet. My reward was a glorious panoramic view, in- 
cluding snowy passes towards Duber and Chihil-dara, and 
I found abundance of edelweiss lower down, just appearing 
under the influence of the first sunny spring days. The 
descent to some almost completely decayed ruins led by a 
goat track down a side of the mountain so steep and so slip- 
pery that if previous inspection had been possible I might 
well have felt tempted to leave them alone. 

On April 24th, after recording specimens of the Dardic 
dialect spoken by some Duber men whom Shah *Alam had 
collected, I started down the valley. The quest of Aornos 
was now drawing me eagerly to the south, towards the far- 
stretching range that divides the drainage of the Ghorband 
river and its chief tributary, the river of Kana, from the 
hill tract of Chakesar. It was on the eastern extremity of 
this range that the long-sought-for site was most likely to 
be found. But for the attraction of this goal I should gladly 
have spared a little more time for Kana ; so bright was the 
scenery of this fertile open valley and so cheering the hearty 
welcome of its Khans and people. Even the many roadside 
graves gaily decked with big irises were a pleasure to 
behold (Fig. 68). 

Kana, occupied by the Jinki-khel clan in its upper and 
main portion, differs in several ways from other tracts of 
the Indus Kohistan to which the tide of Pathan invasion 
has extended. It does not know that mischievous system 
of wesh, or periodical redistribution of all lands, which else- 
where in this region prevents the undertaking of any 
permanent improvements. Kana had not, indeed, escaped 



no THE SHILKAI PASS AND KANA CH. xv 

that bane of local feuds and vendettas to which the wesh 
practice contributes so largely. The numerous ruined or 
deserted dwellings that I saw at Bilkanai were visible proof 
of this. Yet somehow the people of Kana had inherited a 
sense of feudal attachment to their Khans, and this has 
evidently helped a good deal towards a ready acceptance 
of the Badshah's new regime. 

The two Khans of Kana, Amir Khan and Pirdad Khan, 
had been intelligent enough to attach themselves to the 
cause of the Miangul long before he gained overlordship. 
When in 1922 the Pathan clans east of Swat formed a con- 
federacy to resist the Miangul's 'forward policy' and gave 
battle to his invading force in the Mukhozai country, the 
Khans kept aloof, though they could not prevent their 
tribesmen from sharing in the fray and defeat. They had 
themselves been besieged and had suffered losses. But with 
the success of the Miangul they received their reward. 
Everywhere else he had caused forts and towers of local 
headmen to be razed to the ground, and had imposed his 
own trusted servants to manage affairs and enjoy the sweets 
of office. But his friends, the Khans of Kana, had been 
allowed to keep their forts and to exercise their old quasi- 
baronial powers in the person of their sons, who were 
appointed 'Tahsildars'. 

Naturally enough, both Amir Khan and Pirdad Khan 
were eager to demonstrate their loyalty by a special display 
of hospitality. So as I travelled down Kana we were every- 
where treated to gay gatherings of their armed retainers, 
with fluttering banners and much martial music on drum 
and bagpipe (Fig. 70). It was impossible for me to do 
due justice to all the festivities prepared at Bar-kana, Delai, 
Nawe-kile. But my Swati protectors and the escort had 
ample occasion on the way to show their receptive 'capacity'. 
The quite antique-looking succession of servitors with trays 



CH. xv BARONIAL HOSPITALITY in 

of viands, large cakes, and sweets could scarcely keep pace 
with the rapid disposal of their supplies by the big circle 
of feasting men. 

Large-hearted hospitality is one of the best traits in 
Pathan character, and there could be no doubt about the 
satisfaction with which old Amir Khan, a venerable grey- 
beard, pressed it upon us at Bar-kana, until the threat of 
rain forced us to make a timely start for our camping-place 
lower down under the shelter of the Badshah's new castle 
at Nawe-kile. What pleased me most was the tasteful 
decoration of the places where we met with this hospitable 
reception. Glowing red rhododendron blossoms and big 
white irises were hung on the shrubs and saplings that en- 
closed them. The rhododendrons were brought down from 
the mountains, the irises from the burial-grounds on which 
they grow as abundantly as in Kashmir. 

During my visit to Kana I enjoyed, perhaps more than 
anywhere else on this expedition, what seemed like a bodily 
translation into an earlier phase of human society and life. 
It is difficult to express clearly in a few words the effect, 
for one endowed with historical instincts, of close and con- 
stant contact with men whose ways of thought and action 
reflect conditions that the West has left centuries behind. 
To be able to observe these, unconsciously as it were and 
undisturbed by the influences of our complex modern life, 
seemed to me a diverting experience, and at the same time 
a very instructive lesson in history. The ease with which, 
under a quasi-medieval regime, all things really needful 
could be arranged for the favoured few was exemplified in 
my own case. The advantages thus enjoyed were great 
indeed and might well explain and excuse a certain reluc- 
tance on my part to return to 'civilization' and all its 
exactions. 

By the afternoon of April 25th we had reached the lower 



112 THE SHILKAI PASS AND KANA CH. xv 

end of the valley of Kana. We had to take leave of it by 
two slightly trying river crossings. Both the river of Kana 
and that of GhSrband into which it empties itself near 
Karorai are far too deep and rapid to be forded without 
serious risk. They are bridged both above and below the 
confluence where rocky defiles confine them. But in both 
places the bridge actually available consisted of a single 
big rafter, sagging and swaying a good deal (Fig. 72). 
It was, I confess, with some relief that I found myself 
safely on that side of the Ghorband river which is reckoned 
in the tribal area of the Azi-khel centred at Chakesar. Our 
route then took us along the steep slopes of the valley which 
leads straight south up to the range running from the Swat 
watershed to the Indus, and after a long ride we reached 
the village of Upal by nightfall. The 'Subahdar' or com- 
mandant of Upal met us on the way up to his fort with a 
large posse of his men-at-arms, who marched, with flags 
flying, to the weird music of a drum and a bagpipe and 
enlivened the last stage of our march. 



CHAPTER XVI 

THE ASCENT TO PlR-SAR 

IT was from Upal (Fig. 74), some 5,000 feet above sea- 
level, that I was to begin the search for a likely site of 
Aornos. The move necessary for this purpose, along the 
crest of the range, could only be made on foot. So the 
mules which had brought our baggage from Kana were 
discharged and our riding ponies dispatched across the 
Upal pass to await us at Chakesar. I was pleasantly sur- 
prised at the ease with which the hundred odd carriers re- 
quired for our combined camp were collected by the morning 
of April 26th. But occasional labour of this kind forms 
part of the conditions of tenure upon which the Gujar and 
other 'Fakirs' are allowed to cultivate the land owned by 
their Azi-khel masters, and under the prevailing feudal 
conditions the latter do not seem to find any difficulty in 
enforcing their claim. 

It was a day full of eager expectation, but also not free 
from anxious uncertainty. On the two preceding nights I 
had carefully read over the descriptions that Arrian and 
Curtius have left us of Alexander's great exploit at Aornos. 
These are detailed enough as regards incidents of the opera- 
tions by which the formidable rock fastness was captured. 
But apart from references to the vicinity of the Indus the 
general topographical indications seemed singularly want- 
ing in precision. Descriptions, in the case of Curtius 
rhetorically elaborated, of the extent and height of the 
mountain stronghold and of the difficulties that the attack- 
ing Macedonians encountered, could not make up for the 
lack of exact distances and bearings from definitely known 
points. 

How I wished that the great Alexander had brought in 
his train some man of letters, accustomed, like a Chinese 



114 THE ASCENT TO PlR-SAR CH. xvi 

annalist, to look out for plain topographical facts, such as 
bearings, distances, and the like, and to record them with 
impartial clearness alongside of important military events! 
If he did, we must regret that the fortunes of classical 
literature have not allowed that record to come down to us. 
The Best I could hope for was to find a ridge or plateau 
which by its size, position, and configuration would account 
for the character of the operations that brought about the 
conquest of the 'Rock'. 

All the more grateful did I feel when what proved a long 
day of physical toil soon brought an encouraging omen. 
Some 'Kapur kandare', ruins of heathen (Kafir) times, had 
been reported by the greybeards of Upal on a little spur 
clearly visible from the village and about 1,000 feet above 
it. I therefore soon left the stony ravine in which our 
long baggage column was moving towards the summit of 
the Upal range as for shortness* sake I may call it and 
with my 'handy-man', the ever-agile Shah 'Alam, and the 
escort, climbed up the spur of Chat. 

I was rewarded for this detour; for the little plateau on 
the top showed, not merely the well-cultivated fields of 
nearly a dozen Gujar households, but also the remains of 
a small circumvallation which by its characteristic masonry 
of the 'Gandhara type* could at once be recognized as dating 
from the Buddhist period. Search in the fields close by 
soon brought to light fragments of decorated pottery of the 
kind which during the preceding weeks I had learned to 
associate with ancient structures in these parts. Neverthe- 
less, it was a pleasant surprise when Mahmud, an intelligent 
young potter from Upal, who was acting as our guide, 
picked up under my eyes from below the ruined wall a well- 
preserved bronze bracelet of unmistakably antique shape 
and showing a snake head. The rain of the preceding days 
had loosened the ground and thus prepared the discovery. 




73 CKLSl Or L PAL RANGE SL1-N FROM CU\T. 




CH. xvi FIRST MENTION OF MOUNT ONA 115 

The reward I paid out on the spot, as I always did in 
such cases, made Mahmud beam with joy. But he was, of 
course, not aware that the ounce of silver was intended also 
as a return for an important piece of information that he 
had given me in his talk as we climbed up to Chat. I had 
been cautiously testing his knowledge about localities 
farther east on the Upal range and in particular about the 
high ridge of Pir-sar, which the Miangul's people, in re- 
sponse to inquiries started by the Political Agent on the 
Malakand since I first proposed my tour in search of 
Aornos, had pointed out as most nearly answering to the 
broad outlines given them. 

Of course Mahmud was quite unaware that the great 
Sultan Sikandar had ever come to these parts; nor did my 
repeated careful inquiries among local Pathans, Gujars, 
and Mullahs reveal the slightest indication that folk-lore or 
quasi-learned tradition in this region in any way connected 
Swat and the adjacent hill tracts with the exploits of Sikan- 
dar, the 'two-horned', the legendary hero of the 'Alexander 
romance' in its Muhammadan version. But in the course 
of my talk with humble but sharp-witted Mahmud I heard 
for the first time the name of Mount Una mentioned. It 
was believed by all people, so he said, to be the highest peak 
on the range that stretches from the pass of Upal to the 
Indus, and just below it in the direction of the great river 
lay Pir-sar, a big alp cherished by the local Gujars as the 
best of all their 'bandas' or summer settlements both for 
grazing and cultivation. 

It did not take long for my philological subconsciousness 
to realize that Ona (pronounced with that peculiar cerebral 
n which represents a nasal affected by a preceding or follow- 
ing r sound) would be the direct phonetic derivative to 
be expected, according to strict linguistic evidence, from 
the Dardic or Sanskrit name that Greek tongues had 



n6 THE ASCENT TO PlR-SAR CH. xvi 

endeavoured to reproduce by Aornos. But, of course, such 
philological indications could have their weight only if the 
actual topographical conditions at Pir-sar were found to 
agree with those details concerning the siege of Aornos 
which have been handed down to us from reliable 
sources. 

The eager wish to reach Pir-sar, if possible, the same day 
made me think little of the steep climb of some 2,000 feet 
that carried us from Chat to the crest of the range (Fig. 
73). For its upper half it lay through a fine forest of 
conifers and ilex mixed, with violets and other little mess- 
engers of early spring strewing the ground in profusion. 
We reached the narrow crest of the range at an elevation 
of close on 8,000 feet, just below the wooded cone whose 
height had been shown as 8,360 feet by the triangulation 
effected from across the Indus in the course of the Black 
Mountain expedition of 1892. The sylvan scenery along 
the knife-edge crest, which fell away in places in sheer 
cliffs, was delightful. My eyes feasted once again on the 
great panorama of snowy mountains to the north, including 
the noble Mankial peaks far away on the horizon (Fig. 
75). What with the enjoyment of this charming alpine 
scenery and the search for a place where the trees would 
allow photographs to be taken to advantage, the onward 
march was naturally delayed. 

It proved far longer and more trying than I had foreseen. 
The track, if such it could be called, led along the rocky 
and precipitous southern face of the range and kept close 
to its crest with all its ups and downs. I was glad when at 
last after much scrambling I caught the first view of the 
Indus valley through the forest. Above it to the south-east 
I could see the snow-topped heights of the Black Mountains, 
and far away, beyond a long succession of boldly serrated 
minor ranges, I could faintly make out the long, undulating 



CH. xvi ALONG CREST OF UPAL RANGE 117 

outlines of Mahaban. How they revived recollections of 
my short raid in 1904, which had disproved the identifica- 
tion of that mountain with Aornos! Once again kindly 
Fate had allowed me to search for the famous 'Rock'. 
Would it this time grant success? 

It was 6 p.m. when we passed a very fine spring gushing 
forth from the side of the Acharo-sar peak, and after 
scrambling for another half-mile we reached the open crest 
of a rocky spur that descends straight from Mount Uija. 
Just below us spread the grazing alp of 'Little Una'. And 
at this point we came in sight of the bare rocky peak of 
Una-sar or 'Mount Una', 8,720 feet high according to the 
triangulation carried out on the last Black Mountain ex- 
pedition. Stretching away from it southward the long and 
flat- topped ridge of Pir-sar now came into view (Fig. 76). 
It was a very striking sight, this long and almost level 
ridge, as it rose there, girt all round with cliffs, above 
precipitous smaller spurs and steep ravines which were seen 
to run down to the Indus, over 5,000 feet below. Pir-sar 
seemed near enough as I looked across the deep valley, 
flanked by precipitous slopes mostly bare of tree growth, 
that separated us from it. But in the end it took us fully 
three hours more to reach it. 

First we had to get past the steep massif of Mount Una. 
Its southern face falls away lower down in sheer walls of 
rock (Fig. 81). We had, therefore, to ascend by a troublesome 
track following the spur that we had just struck at 'Little 
Uija'. Climbing along its crest we reached at last a small 
shoulder, some 200 feet below the actual summit, where it 
was possible to take to the northern well-wooded side of the 
mountain. Here the descent began, almost as tiring, along 
cliffs and steep gullies where the snow still lay thick. It 
brought us to the small tree-girt alp of Burimar (Fig. 78), 
where we found some summer huts of Gujar graziers and 



n8 THE ASCENT TO PlR-SAR CH. xvi 

the fenced-in resting-place of some Muhammadan saint 
under fine trees. 

At first Burimar, as sighted from a distance, had seemed 
to link up with the wooded conical height marking the 
northern end of the long ridge of Pir-sar. But now as we 
passed down the gently sloping grassy alp to its lower edge, 
I discovered that a deep and precipitous ravine, previously 
masked by close tree growth, still separated us from that 
height. It was a surprise, far from pleasant at first sight, 
for tired men, now that dusk was rapidly coming on. But 
as under Mahmud's guidance we stumbled down among 
rocks and pines to the bottom of that narrow ravine, fully 
600 feet as it proved below the alp, a thought soon began to 
cheer me. Was this not the deep gap on Aornos which at 
first baffled the Macedonian attack, after Alexander had 
joined the detachment sent ahead under Ptolemy, the son 
of Lagos, near the top of the mountain? There was no 
time for me then in the growing darkness to examine the 
ground with reference to the expedient by which, after days 
of toil, Alexander had managed to bring his slingers and 
archers sufficiently near the opposite side to reach the de- 
fenders with their missiles. When we arrived at last at the 
bottom of the gully it proved to be a very confined saddle, 
less than forty yards long and only some ten yards across. 
Fallen trees encumbered it and also lay thick in the narrow 
ravines descending on either side. Then my thoughts 
naturally turned to the 'mound' that Alexander is said to 
have had constructed with cut trees and stakes across the 
gap in order to render the assault possible. 

We were now at last arrived below the high northern 
extremity of Pir-sar itself (Fig. 79) nor had we to fear 
the stones and other missiles which an enemy holding the 
heights would have found it so easy to hurl down. Yet the 
ascent past the precipitous cliff's lining the south-western 




76. PlR-SAR RIDGE SLEX FROM SOU! H-WLSTERX SLOPE OF LXA-SAR PEAK. 




CH. xvi CLIMB PAST CLIFFS OF BAR-SAR 119 

slopes of Bar-sar, 'the top hill', as this end of the Plr-sar 
ridge is known, was exacting enough for the fatigue of it 
to be vividly imprinted on my memory. Fortunately to- 
wards the end the moon rose sufficiently high to make foot- 
holds less difficult to find. But I was heartily glad when at 
last by 9 p.m. we reached open ground where the flat portion 
of the ridge adjoins Bar-sar. It was a strange sensation to 
pass next over nearly a mile of practically level ground, 
where the full moon shining under a cloudless sky showed 
verdant fields of young wheat and barley. It served to 
impress me still more with the remarkable natural features 
to which this ridge owes its present attractions fortheGujars 
and which would account for its fame as an ancient moun- 
tain refuge. 

In the centre of the long flat plateau I found our big camp 
pitched near a rudely built mosque, at an elevation which 
subsequent observations showed to be fully 7,100 feet above 
sea-level. The escort and carriers were gathered round big 
bonfires, glad of their protection from the bitterly cold wind. 
It was long after midnight before I could seek warmth 
amidst my rugs. But the growing conviction that Aornos 
was found at last kept my spirits buoyant in spite of be- 
numbed hands and weary feet. Alexander, so Arrian and 
Curtius tell us, offered sacrifices to the gods when he had 
gained possession of the 'Rock'. I had no victory to give 
thanks for. Yet I, too, felt tempted to offer a libation to 
Pallas Athene for the fulfilment of a scholar's hope, long 
cherished and long delayed. 



CHAPTER XVII 

IN SEARCH OF AORNOS 

IT was the search for Aornos that had led me to Pir-sar. In 
order to explain the reasons that made me anxious to 
examine this ground by the Indus with a view to the loca- 
tion of the famous rock fastness, I must turn back to the 
account of Alexander's campaign where we left it. We have 
seen that after the capture of Ora and the subsequent aban- 
donment of Bazira, now confidently located at Blr-kot, 
Alexander's operations in the Swat valley were concluded. 
Having placed garrisons in these strong places to guard the 
country, he turned south to rejoin that large division of his 
army which under Hephaistion and Perdikkas had preceded 
him down the Kabul river into the open Peshawar valley. 
Before relating this junction of the Macedonian forces and 
Alexander's subsequent move to the Indus, Arrian, here as 
elsewhere our chief authority for all that concerns the great 
conqueror's campaign, tells us that on hearing of the fall 
of Ora, the other Assakenoi, >. e. the people of Swat, all left 
their towns and 'fled to the rock in that country called Aornos'. 
Arrian then proceeds to inform us of the reason why 
Alexander was filled with the eager desire to capture that 
rock fastness. His statements on this point, apart from the 
topographical indications that they contain, are of general 
interest for the historical student; for they help to throw 
welcome light on certain psychological factors that un- 
doubtedly played an important part in more than one of 
Alexander's wonderful enterprises just as they did in 
those of his modern counterpart, Napoleon. At the same 
time those statements furnish a significant illustration of 
the critical standpoint from which Arrian was apt to view 
the fabulous element fostered by the hero of his story. This 
is what he tells us of Aornos (iv. xxviii). 



CH. xvn ARRIAN'S ACCOUNT OF AORNOS 121 

This is a mighty mass of rock in that part of the country, and 
a report is current concerning it that even Herakles, the son of 
Zeus, had found it to be impregnable. Now whether the Theban, 
or the Tyrian, or the Egyptian Herakles penetrated so far as to 
the Indians I can neither positively affirm nor deny, but I incline 
to think that he did not penetrate so far ; for we know how common 
it is for men when speaking of things that are difficult to magnify 
the difficulty by declaring that it would baffle even Herakles him- 
self. And in the case of this rock my own conviction is that 
Herakles was mentioned to make the story of its capture all the 
more wonderful. The rock is said to have had a circuit of about 
200 stadia, and at its lowest elevation a height of n stadia. It 
was ascended by a single path cut by the hand of man, yet 
difficult. On the summit of the rock there was, it is also said, 
plenty of pure water which gushed out from a copious spring. 
There was timber besides, and as much good arable land as re- 
quired for its cultivation the labour of a thousand men. 

Alexander on learning these particulars was seized with an 
ardent desire to capture this mountain also, the story current 
about Herakles not being the least of the incentives. 

We may never know whether the ambition stimulated by 
such reports about Aornos was the sole incentive that 
decided Alexander to effect its capture. The decision was 
probably due quite as much, if not more, to the strategic 
principle invariably kept in view by Alexander of never 
leaving an enemy behind him until he had been completely 
crushed. Anyhow, we have seen that instead of pursuing 
the fugitive Assakenoi to their mountain retreat, Alexander 
moved from Swat into the Peshawar valley. There he 
organized the Macedonian control over this important dis- 
trict and then proceeded to the Indus. 

Arrian does not explicitly tell us that Aornos was situated 
on the left bank of the Indus. But the narratives of 
Diodorus and Curtius agree in distinctly indicating this 
position, and there are strong reasons for looking for 
Aornos there. I have set them forth fully elsewhere, and 
a brief summary will here suffice. In the first place a glance 
at the map will show that with Macedonian posts established 



122 IN SEARCH OF AORNOS CH. xvn 

up the main Swat valley as far as the vicinity of Mingaora, 
the bulk of the fugitive population evacuating the towns 
farther up the valley could seek safety neither to the west 
nor to the south. 

In the former direction the way was obviously barred by 
the invaders. To the south, as far as it could be reached by 
routes not commanded by the Macedonian posts guarding 
the main valley, there lay Buner, a country singularly open 
for the most part and accessible by numerous passes from 
the Peshawar valley. This had already been reached by 
the major portion of Alexander's army, and thus Buner, 
too, lay open to invasion. Turning to the north, no safe 
refuge from invasion could be hoped for in that portion of 
the main Swat valley, which, as we have seen, continues 
remarkably open and easy as far as Churrai, and the same 
applies to the side valleys that open from this portion. 
Higher up, in the narrow gorges of Torwal, invasion would, 
no doubt, be kept off by the natural difficulties of the 
ground. But there, no less than at the alpine heads of the 
valleys descending from the high watersheds towards 
the Panjkora and Indus, local resources would have been 
far too limited for the maintenance of a great host of 
fugitives. 

Conditions were far more favourable for a retreat to the 
east. Here the large and for the most part fertile tracts of 
Ghorband, Kana, Puran, Mukhozai, stretching down to 
the Indus, could be reached by several easy passes, open 
throughout the year, leading across the watershed from the 
Swat side to Ghorband. By crossing the range towards 
the Indus the fugitives would place a natural barrier be- 
tween themselves and the invaders. Behind it they would 
find local resources amply sufficient for their maintenance 
until the danger had passed. And, finally, with secure access 
to the Indus they would have there the advantage of being 



CH. xvii RETREAT FROM SWAT TO INDUS 123 

able to draw help from across the river or else to continue 
their retreat in that direction. 

The great importance of this advantage becomes obvious 
when we remember what Arrian has previously told us 
about the help which the defenders of Ora had expected 
from Abisares, the chief whose power extended over the 
present Hazara and adjacent valleys to the left bank of the 
Indus where it faces the lower portions of the tracts above 
named. There is plenty of historical evidence down to quite 
modern times to attest the close relations between Swat and 
this territory to the east of the Indus. The population of 
Hazara is largely composed of a tribe still known as Swatis, 
descended from the pre-Muhammadan inhabitants of Swat 
whom the Pathan invasion of the late Middle Ages had 
driven out of their original seats. Equally significant is the 
fact that the Pathan clans now settled on both the Swat and 
Hazara sides of the river, and closely allied in descent, have 
always shared in the fighting which attended the several 
Black Mountain expeditions since the British annexation 
of the Panjab. 

Thus we have converging evidence to explain why the 
retreat of the inhabitants of Upper Swat had led them east- 
wards to the Indus. At the same time it helps us to under- 
stand the sound strategic reasons which caused Alexander, 
before attacking Aornos, first to turn south to the Peshawar 
valley. Once he had consolidated his hold there and made 
his arrangements for crossing the Indus quite secure, he 
could safely move up its right bank and attack the mountain 
retreat of the Swat fugitives from the south. He thus 
avoided the entanglement in a mountainous region that 
would have attended and hampered direct pursuit from the 
Swat side. The fugitive host could be cut off from retreat 
to the east of the Indus and from such assistance as Abi- 
sares, the ruler on that side, might offer. Finally, when 



124 IN SEARCH OF AORNOS CH. xvn 

attacking Aornos from the south, Alexander could com- 
mand all the advantages which the Indus valley and the 
fertile plains of the Peshawar valley would offer in respect 
of supplies and other resources. 

The importance of this last consideration is clearly brought 
out by what Arrian tells us after briefly recording that 
Alexander, having put a Macedonian garrison into the city 
of Peukelaotis (Sanskrit Pushkalavati), the ancient capital 
of Gandhara, located at the present Charsadda north-east of 
Peshawar, moved eastwards and 'reduced other towns, some 
small ones situated on the Indus'. 

After he had arrived at Embolima, which town lay not far from 
the rock of Aornos, he there left Krateros with a portion of the 
army to collect into the town as much corn as possible and all 
other requisites for a prolonged stay, in order that the Macedonians 
having that place as a base might by protracted investment wear 
out those holding the rock, in case it were not taken at the first 
assault. He himself taking with him the archers, the Agrianians, 
the brigade of Koinos, the lightest and best armed from the rest 
of the phalanx, two hundred of the companion cavalry and a 
hundred mounted archers, marched to the rock. 

Arrian does not indicate the exact position of Embolima. 
But since we have seen that Aornos was situated on the 
right bank of the Indus, the town chosen to serve as Alex- 
ander's base of supplies may with good reason be also 
looked for there. The mention made in Ptolemy's Geography 
of Embolima as a town of Indo-Scythia situated on the 
Indus confirms this. But neither the Greek geographer's 
notice nor the mention of the same place under its Sanskrit 
name Ambulima in that quaint Buddhist text, the Geo- 
graphy of the Demons, to which I had occasion to refer 
in connexion with the blankets of Churrai, helps us to 
determine the exact position of Embolima. 

When General Abbott, the valiant Frontier Officer from 
whom Abbottabad, the administrative centre of Hazara, 



CH. xvii POSITION OF EMBOLIMA 125 

takes its name, discussed in 1854 his location of Aornos 
on the Mahaban range south of Buner, he proposed, as 
M. Court, one of Ranjit Singh's French generals had done 
before him in 1839, to recognize Embolima in the present 
village of Amb situated on the right bank of the Indus. It 
lies about eight miles to the east of Mahaban and is the 
place from which the Nawab of Amb, one of the Miangul's 
recent antagonists, takes his title. This identification of 
Aornos with Mahaban had been proposed without a visit 
to the locality then wholly inaccessible to any European 
and was based mainly on what the telescope used from 
a great distance on the Hazara side of the river was believed 
to show. This identification had proved untenable in the 
light of the true topographical features of the supposed site, 
as revealed by the close survey of Mahaban carried out by 
me in 1904. 

This fact did not necessarily invalidate the location of 
Embolima at Amb. But even if it is accepted, though 
resting solely on the identity of the modern name with the 
first syllable of Embolima philologically weak evidence 
since the apocope of fully three syllables at the end is diffi- 
cult to account for we are still left free to look for Aornos 
higher up on the Indus; for Arrian's narrative shows that 
it took Alexander two marches from Embolima to reach 
the neighbourhood of Aornos. Already, when recording 
the conclusive evidence furnished by my visit to Mahaban 
in 1904 against the conjectured location of Aornos on that 
range, I had realized that we must look for the site farther 
up the great river. But as that tribal territory remained 
inaccessible, it was not until fourteen years later that my 
attention was definitely drawn to an area where a probable 
solution of the problem might be looked for. 

A very valuable clue was then furnished by my friend, 
the late Colonel R. A. Wauhope, R.E., a highly accom- 



126 IN SEARCH OF AORNOS CH. xvn 

plished officer of the Survey of India, whom the Great War 
had temporarily brought back from retirement to the Tri- 
gonometrical Survey Office at Dehra Dun. Work was then 
proceeding on the maps reproducing the surveys carried out 
on my three Central- Asian expeditions. Colonel Wauhope 
had been in charge of the survey operations conducted on 
the left bank of the Indus during the Black Mountain cam- 
paigns of 1888 and 1891-2. From high survey stations 
then established on the Black Mountain range and again 
during the brief occupation of the tribal tracts of Nandihar 
and Allah! farther north, he had become familiar with the 
general features of the Indus valley below and the hills 
overlooking it on the opposite side, all the way from above 
Amb to Chakesar. Remaining a sound classical scholar 
all though his life, he was interested in the question of 
Aornos. From his observations at that time he had come 
to conclude that a position corresponding to that described 
by Alexander's historians was more likely to be found on 
the mountain spurs descending steeply to the Indus opposite 
to the mouth of the Nandihar valley near ThakSt than 
anywhere else. But his experience as a topographer had 
also told him that only by close examination on the spot 
could a definite location be hoped for. 

From what the available map of the Survey of India 
showed of this region, which had then been roughly 
sketched from a distance, there could be no doubt that the 
spurs referred to by Colonel Wauhope were the easternmost 
finger-like offshoots of the range that trends with a due 
easterly bearing and a total length of close on twenty miles 
from the Swat- Indus watershed to the Indus and divides 
the Ghorband valley from Chakesar. That this range drops 
very steeply towards the river, which flows round its foot 
in a sharp bend, was evident enough from the short distance 
that the map showed between the highest triangulated point 




So. VIEW OF SNOWY R \NGK AT HI AD Or SWAT VALIKY, LOOKING NORTH 
FROM LANDE-SAR. 

High pc.iks ibo\e Manki.'il .ire MVII to nylit 




JSl. rNA-SAR 1'1'AK ShKN FROM KUZ-SAR. 

i 'i, n anrl Arhar alns on slopes to Irft, B.il.u on spur below them; Bunmar on right. 



CH. xvii SPURS SIGHTED BY COL. WAUHOPE 127 

on its crest and the river-bed at Thakot, fully 7,000 feet 
lower. I could gather little else from the map regarding 
the topography of the high ground that we had gained in 
the dark by our trying march to Pir-sar. All the more 
eagerly did I long for the morning which was to reveal it 
in full clearness. 



CHAPTER XVIII 

THE SURVEY OF PlR-SAR 

THE violent gusts of wind that shook my little tent during 
the night of my arrival on Pir-sar left but a poor chance of 
sleep before I rose next morning at daybreak. The view 
that spread before me at this well-detached height of some 
7,100 feet above sea-level was extraordinarily wide and 
grand. But the icy blasts blowing down the Indus from 
the snow-covered ranges of the Kohistan, comparatively so 
near, made it difficult to enjoy it fully until midday had 
brought calm and warmth. It was the same throughout 
the three days that we spent on this exposed height. None 
the less I found the work during these days most fascinat- 
ing, favoured as it was by continuous clear weather. The 
task involved a detailed survey of Pir-sar and the whole 
region around it, supplemented by whatever information 
could be gathered about practicable routes, cultivation, 
water-supply, and other local conditions. 

It was in the course of the close survey unremittingly 
pursued during this period that I realized more and more 
clearly that striking agreement of topographical features 
which brought the conviction that in this remarkable ridge 
of Pir-sar we have indeed the long-sought-for rock fastness 
of Aornos. The sketch-map reproduced from the plane- 
table survey that Surveyor TSrabaz Khan prepared under 
my supervision on the large scale of four inches to the mile, 
together with the photographs (Figs. 76-86), will help to 
illustrate these features. Yet some description of details is 
needed before we proceed to review the records left to us 
of Alexander's siege of Aornos and to examine the relevant 
details of this story in the light of the topographical know- 
ledge acquired. 

Pir-sar is but one of a series of narrow spurs which the 



CH. xvra LEVEL TOP OF SPUR 129 

range stretching east from above Upal throws out to the 
south before it drops rapidly and flattens out fanlike towards 
the low plateau of Maira washed at its foot by the Indus. 
Of these spurs Plr-sar preserves its height for the longest 
distance, and owing to the uniform level and the very fertile 
soil of its summit affords most scope both for cultivation 
and grazing. The practically level portion of the top 
(Fig. 85) extends at an average elevation of about 7,100 feet 
for over a mile and a half. At its upper end this flat portion 
is bordered for some distance by gentle slopes equally suited 
for such use (Fig. 82). 

Owing to its greater height and the depth of the valleys 
on either side Pir-sar forms a dominating position; over- 
looking all the other spurs, it offers an exceptionally wide 
and impressive view. This comprises the whole of the Indus 
valley from below the Mahaban range in the south to where 
the winding course of the great river lies hidden between 
closely packed spurs descending from the high snowy 
ranges towards Kaghan and the Swat headwaters (Fig. 87). 
To give some idea of the extent of the vast panorama 
commanded from Plr-sar I may mention that it includes 
northward the great ice-crowned peaks above T5rwal, 
Duber, and Kandia (Fig. 80), and to the east all the ranges 
that adjoin the central part of Hazara. Southward the 
plain of the Peshawar valley above Attock could be dis- 
tinctly seen. 

The spur from its level top, to which the name Pir-sar, 
'the holy man's height', is properly applied, falls away both 
on the east and west in very steep rocky slopes (Fig. 77). 
In places these form sheer cliffs, while in others pines and 
firs have managed to secure a footing. The southern end 
of Plr-sar rises into a small but conspicuous hillock, known 
as Kuz-sar, 'the lower height', as opposed to the Bar-sar at 
the northern end (Fig. 82). There the spur divides into 

K 



130 THE SURVEY OF PlR-SAR CH. xvin 

three narrow branches, all flanked by precipitous rocky 
slopes (Fig. 83). The crest of the middle one is in its 
upper portion so steep and narrow as to be practically in- 
accessible. The shortest branch, called Maju, juts out like 
a bastion to the south-west, before it terminates in sheer 
cliffs at^a level of about 1,600 feet below the top of Pir-sar. 

The western slope of Pir-sar descends steeply for some 
2,000 feet into a very confined valley (Fig. 77). The bottom 
of this forms in parts an impracticable ravine, while in others 
little terraces bear a few scattered fields. On the opposite 
side of the valley rises the small spur of Balai, flanked by 
formidable bare cliffs, almost perpendicular in places. There 
are short stretches of more gentle slope on its summit, which 
are used for summer grazing; but these are practically 
accessible only from the crest of the main range just below 
the Oija-sar peak. A deep ravine divides the spur of Balai 
westwards from another and much longer one, known to 
the local Gujars as Danda-Nurdai. This detaches itself 
from the main range near the grazing-grounds of Landai 
and farther down faces the south-western slopes of Pir-sar. 

Its narrow serrated crest is crossed by two passes. The 
lower one, called Pezal-kandao, with an elevation of about 
4,000 feet, gives access to a portion of the valley where 
opposite to the cliffs of Maju some cultivation is carried on 
by the scattered homesteads of the Gujar hamlet of Talun. 
From below the Pezal-kandao it is possible to ascend by a 
difficult track to the crest of the Maju spur, and thence to the 
southern end of Pir-sar. Across the other pass, about 6,500 
feet above sea-level, a somewhat easier route leads from 
the valley behind the Danda-Nurdai spur to the grassy 
slopes below the alp of Little Uija, and thence joins the 
track passing along the top of the main range. We shall 
see below that these passes may claim some interest in con- 
nexion with the story of Aornos. 




83. CLIFFS BELOW KVZ-SAR END OF PlK-SAR, SEEN FROM 
ASHARA1 RIDGE. 



CH. xviii ADJACENT RIDGES AND VALLEYS 131 

From here we must turn back to Plr-sar to acquaint our- 
selves rapidly with the ground which adjoins it to the east. 
That it differs in some respects from that observed to the 
west is due mainly to the fact that the main range, after 
throwing off to the south the commanding spur of Pir-sar, 
very soon falls off in height and becomes bare of trees. The 
drainage here gathers in one wide trough before taking its 
course to the Indus. Between the deeply eroded Nullahs 
that join this trough there rises a succession of short knolls 
and ridges. All have very steep slopes, but are crowned by 
little plateaux which, as seen from Pir-sar, give them an 
appearance curiously suggestive of small detached islands 
(Fig. 84). Some of these little hill-tops bear patches of 
cultivation; but all are devoid of trees and water, and only 
capable of temporary occupation. The slopes of Plr-sar 
facing east also descend very steeply. 

It only remains to describe briefly the summit of the Plr- 
sar spur. This presents itself for a distance of a little over 
a mile and a half as an almost level plateau, occupied along 
practically its whole length by fields of wheat. The width 
of the cultivated ground on the top varies from about 100 
to 200 yards, with strips available for grazing by the side 
of the fields. Fine old trees form small groves in places 
(Fig. 85), and one of these near the middle of the ridge 
shelters a much-frequented Ziarat, or shrine. There are 
several small springs in the little gullies that furrow the 
steep slopes close below the ridge, and these feed the streams 
that drain into the valleys below. But, in addition, two 
large reservoirs have been constructed of bands of rough 
stonework to store a large supply of water from rain or 
melting snow, and thus to meet the need of the herds of 
cattle that are brought to graze here during the summer 
months. We found them filled to a depth of several feet. 

Over two dozen homesteads, roughly built in the Gujar 

K2 



132 THE SURVEY OF PlR-SAR CH. xvm 

fashion, and scattered in groups over the plateau, serve to 
shelter the families that move up with their cattle and 
occupy Pir-sar from late spring till the autumn. The 
mosque that I shall presently refer to forms the centre of 
the settlement. Owing to the fact that the Pir-sar ridge 
stretches from north to south and is nowhere shaded by 
higher ground, its summit receives an abundance of sun- 
shine. It consequently becomes clear of snow very early 
in the year. This explains also why, in spite of an abnor- 
mally late spring, we found the wheat already stand- 
ing high. 

At its southern end Pir-sar is guarded, as it were, by the 
hill of Kuz-sar already mentioned, which rises about 100 
feet above the plateau and completely commands the diffi- 
cult paths leading up from the Maju and Asharai crests. 
At the northern extremity the plateau is still more effec- 
tively protected by the bold conical hill of Bar-sar, which 
rises to a height of about 7,900 feet, and is thus at its 
summit about 800 feet higher than the plateau. The ap- 
proach from the latter to the thickly wooded top lies first 
over easy grassy slopes (Fig. 82), but for about the last 
300 feet becomes very steep and rocky. The top portion 
of Bar-sar has a distinctly triangular shape. It is fringed 
with crags along each of its sides and these are very preci- 
pitous except at the angle pointing north. There an easier 
slope leads down 200 feet to a narrow saddle, and close 
beyond it rises a small flat-topped outlier of Bar-sar known 
as Lande-sar, 'the lower height' (Fig. 80). Its elevation 
is but little less than that of Bar-sar, and the slopes below 
it are very steep and rocky. 

It is by the angle pointing west that Bar-sar joins up 
with the main range, in the ^xial line of which it lies. But 
it is just here that the continuity of the range is broken by 
the deep and precipitous ravine that we encountered on our 



CH. xvm SHOULDER OF MASHLUN 133 

first approach to Pir-sar. The bottom of this ravine lies 
approximately on the same level as the plateau of Pir-sar 
and about 600 feet below the alp of Burimar which, as we 
have seen, faces Bar-sar. I have already described the 
troublesome descent from Burimar to the bottom of the 
ravine known as Burimar- kandao. But the angle at which 
the narrow rocky ar&te from the top of Bar-sar runs down 
to it is still steeper. 

The succession of crags, in places almost vertical, is here, 
however, broken at one point by a small projecting shoulder, 
called Mdshlun. This, visible in Fig. 79, is quite flat on 
its top and extends for about half a furlong westwards, 
with a width of some thirty yards at its end. Trees grow 
on it thickly, as they do also on the rocky slopes above and 
below. This shoulder of Mashlun juts out at a height of 
about 450 feet above the bottom of the ravine, and behind 
it precipitous cliffs rise for another 350 feet or so higher to 
the summit of Bar-sar. To the remains of an ancient fort 
traceable on this summit, and to the important topo- 
graphical indication presented by the shoulder of Mashlun, 
I shall presently recur. 

Having now described the actual configuration, I may 
briefly sum up the essential features that necessarily in 
vested it with exceptional advantages as a place of safety 
and natural stronghold for the ancient inhabitants of this 
region. Its great elevation, more than 5,000 feet above the 
Indus, would alone make attack difficult. The extent of 
level space on its top, greater than that to be found on any 
height of equal natural strength farther down on the right 
bank of the Indus, would permit of the assembly of large 
numbers both for safety and for defence. Its central 
position would make Pir-sar a particularly convenient rally- 
ing-place for large and fertile hill tracts such as Chakesar 
and Ghorband, as well as for that portion of the Indus 



134 THE SURVEY OF PlR-SAR CH. xvin 

valley lying close below, where the space available for culti- 
vation is wide and villages accordingly large and numerous. 
The great height and steepness of the slopes with which 
Pir-sar is girt would suffice to make its defence easy in 
times when those fighting from a superior height had every 
physical advantage on their side. And in this respect full 
account must also be taken of the fact that even on the side 
where the spur adjoins and is overlooked by the main 
range, it is isolated by the deep ravine of the Burimar- 
kandao. Nor should the great strategic strength of the 
general position be overlooked, considering that over two- 
thirds of it, as the map shows, are protected by the great 
bend of the Indus. 



CHAPTER XIX 

THE STORY OF ALEXANDER'S SIEGE OF AORNOS 

FROM our survey of Pir-sar we may now turn back to the 
record of Alexander's operations where we left it on his 
arrival in the vicinity of Aornos. Among the extant accounts 
of Alexander's great feat there, that of Arrian (Anabasis, 
iv. xxix-xxx) is the fullest and also undoubtedly the 
most reliable. We may attach all the more value to it in 
the present connexion because, of the two contemporary 
authorities whose narratives Arrian in his preface declares 
to be worthy of more credit than all the rest, one was that 
same Ptolemy, son of Lagos and the first of the Ptolemies 
of Egypt, who personally played a very important part in 
the conquest of Aornos. Arrian's account of the operations 
which led to the capture of the rock fastness of the fugitive 
Assakenoi is so clear and instructive in its topographical 
details that it seems best to reproduce Mr. M c Crindle's 
translation of it in extenso, with a few slight alterations 
which an examination of the original text appears to me to 
render necessary. 

Some men thereupon who belonged to the neighbourhood came 
to him, and after proffering their submission undertook to guide 
him to the place most suited for an attack upon the rock, that 
from which it would not be difficult to capture the place. With 
these men he sent Ptolemy, the son of Lagos, and a member of 
the bodyguard, leading the Agrianians and the other light-armed 
troops and the selected hypaspists, and directed him, on securing 
the position, to hold it with a strong guard and signal to him when 
he had occupied it. Ptolemy, following a route which was trying 
and difficult, secured the position without being perceived by the 
barbarians. He fortified this all round with a palisade and a 
trench, and then raised a beacon on that part of the mountain 
from which it could be seen by Alexander. 

The signal fire was seen, and next day Alexander moved forward 
with his army; but as the barbarians offered valiant opposition 
he could do nothing more owing to the difficult nature of the 



136 ALEXANDER'S SIEGE OF AORNOS CH. xix 

ground. When the barbarians perceived that Alexander had found 
an attack [on that side] to be impracticable, they turned round 
and attacked Ptolemy's men. Between these and the Macedo- 
nians hard fighting ensued, the Indians making strenuous efforts to 
destroy the palisade and Ptolemy to hold the position. The bar- 
barians had the worse in the skirmish, and when night fell with- 
drew. 

From,, the Indian deserters Alexander selected one who knew 
the country and could otherwise be trusted, and sent him by night 
to Ptolemy with a letter importing that when he himself assailed 
the rock, Ptolemy should not content himself with holdinjg his 
position but should fall upon the barbarians on the mountain, so 
that the Indians, being attacked on both sides, might be perplexed 
how to act. Alexander, starting at daybreak from his camp, led 
his army to that approach by which Ptolemy had ascended un- 
observed, being convinced that if he forced a passage that way 
and effected a junction with Ptolemy's men, the work still before 
him would not be difficult. 

And so it turned out; for up to midday there continued hard 
fighting between the Indians and Macedonians, the latter forcing 
their way up while the former plied them with missiles as they 
ascended. But as the Macedonians did not slacken their efforts, 
others succeeding to others, while those [before] in advance rested, 
they gained with trouble the pass in the afternoon and joined 
Ptolemy's men. The troops being now all united were thence put 
again in motion towards the rock itself; but an assault upon it 
was still impracticable. So came this day to its end. 

Next day at dawn he ordered the soldiers to cut a hundred 
stakes per man. When the stakes had been cut he began from the 
top of the height on which they were encamped, to pile up towards 
the rock a great mound, whence he thought it would be possible 
for arrows and for missiles shot from engines to reach the de- 
fenders. Every one took part in the work, helping to pile up the 
mound. He himself was present to superintend, commending 
those that with eagerness advanced the work, and chastising any 
one that at the moment was idling. 

The army on that first day extended the mound the length of 
a stadion. On the following day the slingers, by slinging stones 
at the Indians from the mound just constructed, and the bolts 
shot from the engines drove back the sallies made by the Indians 
on those engaged upon the mound. The work of piling it up went 
on for three days, without intermission. On the fourth day a few 
Macedonians had forced their way to and secured a small hillock 



CH. xix ARRIAN'S STORY OF OPERATIONS 137 

level with the rock. Alexander without ever resting drove the 
mound forward, intending to join the mound to the hillock which 
the handful of men already held for him. 

But the Indians, terror-struck at the unheard-of audacity of 
the Macedonians who had forced their way to the hillock, and on 
seeing the mound already connected with it, abstained from 
further resistance, and sending their herald to Alexander, pro- 
fessed their willingness to surrender the rock if he would treat for 
peace with them. But the purpose they had in view was to con- 
sume the day in spinning out negotiations, and to disperse by 
night to their several homes. When Alexander perceived this he 
gave them time to start off as well as to withdraw the round of 
sentries everywhere. He himself remained quiet until they began 
their retreat; and then he took with him seven hundred of the 
bodyguard and of the hypaspists and was the first to scale the 
rock where it had been abandoned. The Macedonians climbed up 
after him, pulling one another up, some at one place, some at 
another. And then at a preconcerted signal they turned upon the 
retreating barbarians and slew many of them in the flight; some 
others retreating in terror flung themselves down the precipices 
and died. Alexander thus became master of the rock which had 
baffled Herakles himself. 

With this clear, sober, and full record of Arrian the 
accounts given by Diodorus and Curtius agree in all essen- 
tial topographical points. That these two authors used 
common sources here as elsewhere is evident from various 
indications. But Diodorus contents himself with a much- 
condensed abstract, and Curtius' narrative owes its greater 
length mainly to his usual expansion of such minor aspects 
of the story as specially lend themselves to rhetorical treat- 
ment. It will therefore be sufficient, in either account, to 
note only those points which have a bearing on the location 
of Aornos. 

Diodorus (Bibliotheca> xvn. Ixxxv) describes the "Rock* 
as a natural stronghold, 100 stadia in circumference, 16 
stadia in height, and with a level surface forming a complete 
circle. The Indus washed its foot on the south; elsewhere 
it was surrounded by deep ravines and inaccessible cliffs. 



138 ALEXANDER'S SIEGE OF AORNOS CH. xix 

An old man familiar with the neighbourhood promised, 
for a reward, to take Alexander up the difficult ascent to 
a position which would command the barbarians in occu- 
pation of the ' Rock'. Guided by him, Alexander first seized 
the pass leading to the 'Rock', and as there was no other exit 
from it, blocked up the barbarians. He then filled up the 
ravine which lay at the foot of the 'Rock' with a mound, and 
thus getting nearer, vigorously pushed the siege by assaults 
continued without intermission during seven days and 
nights. At first the barbarians had the advantage owing to 
the greater height of their position. But when the mound 
was completed and catapults and other engines had been 
brought into action, the Indians were struck with despair 
and escaped from the 'Rock' at night by the pass from which 
Alexander had on purpose withdrawn the guard previously 
placed there. Thus Alexander secured the 'Rock' without 
risk. 

Curtius (Historiae, vm. xi) in his description of the ' Rock' 
(petra), which he calls by the name of A arms, does not give 
any dimensions but mentions that the Indus, deep and con- 
fined between steep banks, washes its foot. Elsewhere there 
were ravines and craggy precipices. In rhetorical style, 
apparently inspired by a reminiscence of Livy, Curtius 
likens the 'Rock' to the meta of the Roman circus, 'which 
has a wide base, tapers off in ascending, and terminates in 
a sharp pinnacle'. This description, if it is based on some 
passage of his original source, would suggest that one portion 
of the ' Rock' rose into a steep conical point. We are told that 
under the guidance of an old man from the neighbourhood 
a light-armed detachment was sent ahead by a detour to 
occupy the highest summit unobserved by the enemy. 

Curtius next relates that in order to make an assault 
practicable a ravine was filled up with a mound. For this 
the trees of a forest close at hand were cut down and their 



CH. xix ACCOUNTS OF DIODORUS AND CURTIUS 139 

trunks, stripped of branches and leaves, thrown in. By the 
seventh day the hollows had been filled. An assault up the 
steep slopes by the archers and Agriani was then ordered. 
Thirty selected youths from among the king's pages under 
Charus and Alexander formed the forlorn hope. In the 
highly rhetorical description that follows it is, however, 
the king himself who is said to have taken the lead in the 
assault. Many are said to have perished, falling from the 
steep crags into the river that flowed below, 'since the bar- 
barians rolled down huge stones upon those climbing up, 
and such as were struck by them fell headlong from their 
insecure and slippery footing*. A long and poetical account 
is then given of the death of the two leaders, Charus and 
Alexander, who had got up high enough to engage in hand- 
to-hand fighting, but were overpowered and fell. 

The king, affected by these losses, then ordered the re- 
treat, which was carried out in an orderly fashion. Alex- 
ander, though resolved to abandon the enterprise, yet made 
demonstrations of continuing the siege. Thereupon the 
Indians, with a show of confidence and even triumph, 
feasted for two days and two nights, but on the third night 
abandoned the 'Rock'. When their retirement was dis- 
covered, the king ordered his troops to raise a general shout. 
This struck such terror into the fugitives that many, 'flinging 
themselves headlong over the slippery rocks and precipices', 
were killed or were left behind injured. 

The three accounts translated or analysed above are the 
only ones that we possess that furnish any specific data 
about Aornos. From a comparison of them we can deduce 
the following definite indications as regards the locality 
referred to. Aornos was a natural stronghold, situated 
on a mountain of great height, rendered capable of easy 
defence against an aggressor by precipitous rocky slopes 
and deep-cut valleys below it. It is important to note that 



140 ALEXANDER'S SIEGE OF AORNOS CH. xix 

no mention is made anywhere of fortification by the hand 
of man. There was sufficient level space on the top to per- 
mit of considerable numbers finding there a safe refuge. 
The site was near to the Indus, which flowed at its foot. 
Its relative height must have been very striking to 
account for the definite measurements of n and 16 stadia, 
approximately corresponding to 6,600 or 9,600 feet, re- 
corded by Arrian and Diodorus respectively. In the same 
way the circuits of 200 and 100 stadia mentioned by these 
two authors, approximately corresponding to 22 and n 
miles, can obviously apply only to a mountain massif or 
range and not to a single hill or peak. 

That Aornos was situated on such a massif or range is in 
fact made perfectly clear by what all three authors relate 
of the commanding height occupied by the Macedonians 
before the beginning of the siege and reached after an ardu- 
ous ascent. Both Arrian and Curtius state that the march 
by which the light-armed detachment sent ahead by Alex- 
ander secured this position under local guidance remained 
unobserved by the enemy. This distinctly suggests that the 
route followed led up a valley that was hidden from the 
view of the defenders of Aornos. 

This assumption finds strong support in Arrian's refer- 
ence to the pass to which Alexander, when subsequently 
following the same difficult route, had to ascend amidst 
severe fighting, before he could join Ptolemy's detachment 
holding the position above Aornos. Incidentally the op- 
position here encountered by Alexander indicates that this 
route leading to the summit of the range, though not visible 
from Aornos and hence not obstructed on the first occasion, 
was yet accessible to its defenders without their having first 
to dislodge the detachment on the height. We learn from 
Arrian that an attempt to dislodge it had in fact been made 
on the preceding day but had failed. 



CH. xix TOPOGRAPHICAL FEATURES NOTICED 141 

We come now to the most significant among the topo- 
graphical features recorded in connexion with Alexander's 
siege of Aornos: I refer to the deep ravine separating the 
heights on which stood the Macedonian camp from the 
nearest part of the 'Rock'. Here, too, Arrian's account is 
the fullest and clearest. It shows us that the primary object 
for which Alexander had to resort to the expedient of con- 
structing a great mound across the ravine was to bring the 
opposite slope held by the enemy within effective range of 
what by an anachronism might be called the small arms and 
field artillery of his force. The precipitous nature of that 
slope would lend itself to easy and most effective defence, 
in particular by rolling down large stones, a formidable 
method of defence the actual use of which Curtius here 
specially mentions. We find a very striking illustration 
of the results which in modern times have attended this 
means of defence on alpine ground in the achievement of 
the valiant bands of Tyrolese peasants who successfully 
protected their country in 1809 from invasion by Napoleon's 
French and Bavarian troops. No assault could succeed 
against Aornos until 'it would be possible for arrows and 
for missiles shot from engines to reach the defenders'. 

We obtain some indication of the great width of the 
ravine, and indirectly also of its depth, from Arrian's state- 
ments concerning the construction of this mound. By the 
united efforts of the troops it was extended on the first day 
the length of a stadion, i.e. about 600 feet. After this it 
became possible, by means of slingers posted on the mound 
and by shots from the engines, to drive back sallies made 
against those engaged on the construction of the mound. 
But 'the work of piling it up went on for three days without 
intermission' before an assault made on the fourth enabled 
a handful of Macedonians to establish themselves on 'a 
small hill which was on a level with the rock'. 



142 ALEXANDER'S SIEGE OF AORNOS CH. xix 

Yet even after this, we are told by Arrian, the construc- 
tion of the mound was continued until it was joined up with 
the position thus gained. This position must have lain still 
considerably below the crest of the height that faced the 
ravine from the side of the 'Rock'. Thus only is it possible 
to account for the stiff climb which it cost Alexander and 
his selected seven hundred to reach the top and fall upon 
the retreating barbarians during the night following their 
offer of surrender. 



CHAPTER XX 

AORNOS LOCATED ON PIR-SAR 

OUR survey as recorded in the map and the preceding 
description have made it easy to recognize, in the local 
features of Pir-sar and its environs, all the topographical 
details of Aornos as they appear from the account of Alex- 
ander's siege. Taking the general features first, we see from 
the map that the Indus flows in a wide bend round that 
eastern extremity of the range of which the Pir-sar spur is 
the largest and most conspicuous offshoot. The more 
specific statement of Diodorus that the Indus washed the 
rock on its southern side is borne out by the map, which 
shows that the portion of this bend which a force coming up 
the Indus valley would first reach lies due south of Pir-sar. 

The relative elevation of Bar-sar at the northern end of 
the spur (7,914 feet by clinometer), if measured from the 
bank of the Indus (about 1,700 feet at Thakot), agrees re- 
markably well with the height of Aornos, 1 1 stadia or about 
6,600 feet, as recorded by Arrian. If the relative height of 
the Uiia peak (8,721 feet above sea-level by triangulation) 
is taken, the agreement becomes, if anything, still closer. 
Obviously no such test can be applied to the measurement 
of the circuit, for we do not know on what lines or on which 
level it was taken. It is curious to note that if a map 
measurer is passed round the foot of the eastern extremity 
of the range from near Sarkul on the Indus past the Takhta 
pass to Shang and thence back again behind the Cija peak 
we get a total direct length of some twenty-two miles. But, 
of course, other measurements, greater or lesser, would also 
be possible. 

Coming next to the commanding height near Aornos 
which a light-armed force was sent ahead under Ptolemy 
to occupy, it is clear that the small plateaux on either flank 



144 AORNOS LOCATED ON PlR-SAR CH. xx 

of Mount Ui?a would exactly answer the purpose in view. 
This was to secure a position on that side from which the 
'Rock 1 was most assailable. Taking into account all the 
tactical advantages that the possession of higher ground 
must have given the assailant, before the invention of long- 
range fire-arms even more than in later times, there can be 
no doubt that the side whence an attack upon the rock-girt 
plateau of Pir-sar would offer most chances of success would 
be where the spur was attached to, and overlooked by, the 
main range. This is the Burimar plateau on the eastern 
shoulder of the culminating peak of Uija-sar (Fig. 78). 

But there are considerations which incline me to favour 
the gently sloping alp of 'Little Ona' immediately below 
the western flank of Ona-sar as the most likely site of 
Ptolemy's fortified encampment. From here it was easier 
to guard the route leading up from the river, and thus to 
give that support to the subsequent ascent of the main force 
which Arrian's account shows to have become indispensable 
once the defenders had discovered the Macedonian move. 
'Little Oija' offers also the advantage, at any rate to-day, 
of easier access to water, and by its situation it was less 
exposed to attack from the enemy's main position on 
Pir-sar. 

The route by which the crest of the range where it over- 
looks Pir-sar could best be gained from the river certainly 
led up the valley to the west of the Danda-Nurdai spur, and 
thence from the head of the valley to 'Little Oija'. The 
information collected by me showed that this route is con- 
sidered the easiest by which the grazing-grounds on the top 
of the main range can be reached from that side. It is 
regularly used by the local Gujars when moving to those 
pastures from their hamlets above the Indus. Near the 
head of the valley the pass shown in the map as having a 
clinometrical height of 6,741 feet gives access to the lower 



CH. xx HEIGHT OCCUPIED BY PTOLEMY 145 

slopes of 'Little Oga', and from these the alps occupied by 
the Gujar huts of Achar and Little Oija can be gained 
without difficulty. 

It is the route just described that I believe, for the reasons 
indicated, to have been followed first by Ptolemy and then 
also by Alexander's main column. Arrian tells us that 
after Alexander had seen the beacon lit by Ptolemy on the 
mountain he had occupied, he next day moved forward 
with his troops, but as his progress was obstructed by 
the barbarians, 'he could do nothing more owing to the 
difficult nature of the ground'. Reference to the map will 
show how easy it was for the enemy collected on Pir-sar to 
obstruct Alexander's march up that valley once Ptolemy's 
preceding move had been discovered and had indicated the 
direction which Alexander's attack was likely to take. The 
valley west of the Danda-Nurdai spur is within easy reach 
from the south-western outlier of Pir-sar across the heigh 
above the pass known as Pezal-kandao, 4,620 feet a 
sea-level. By crowning these heights the enemy c 
seriously interfere with the advance of the Macedonia 
the valley, without risking a battle in the open. It 
equally easy for them, when Alexander's advance up 
valley had been brought to a standstill, to turn round a: 
moving higher up, to attack Ptolemy's detachment holdin; 
the fortified camp, which, as we have seen, may be placed 
at or near Little Oija. 

This attack was beaten off, and when Alexander on the 
next day resumed his advance up the valley, the Indians 
who contested it were attacked in the rear by Ptolemy, to 
whom Alexander during the night had managed to sen 
orders to this effect, as recorded by Arrian. The impo: 
of this help, as well as the difficulties encountered by 
ander, can be well understood by looking at the ma; 
until the pass marked there with the height of 6,741 

I 



146 AORNOS LOCATED ON PlR-SAR CH. xx 

been taken could the junction with Ptolemy's force be 
effected, and considering its elevation and the steepness of 
the Danda-Nurdai spur, Arrian's description of the severe 
struggle required to gain it cannot have been exaggerated. 

Once the Macedonian forces were united in the course of 
the afternoon the further advance towards the 'Rock', which 
Arrian mentions as having been made during the remainder 
of the day, could present no difficulty. This advance would 
necessarily lie along the crest of the range as far as the 
Burimar plateau. That it came to a standstill, as Arrian 
records, without any attack on the 'Rock' being for the 
moment possible, is fully explained by the great natural 
obstacle met beyond, the fosse of the Burimar ravine. 

I have already described the general character of this 
ravine, its considerable depth, and the precipitous nature of 
its slopes. But in order to realize better how fully its 
features explain Alexander's resort to the construction of 
a mound for the purpose of crossing it, attention must be 
called to some details. I have referred above to the pro- 
tection afforded to Pir-sar by the extremely steep rocky 
slopes stretching from the Bar-sar hill, its northern bastion, 
down towards the ravine, some 800 feet below, that separ- 
ates it from Burimar. These slopes, so easily defended from 
above, could not be attacked with any chance of success 
unless they could be brought within the range of missiles. 
Now the direct distance separating the top of Bar-sar 
from ground of approximately the same level on the 
Burimar plateau is some 1,300 yards, and that between the 
Mashlun shoulder of Bar-sar and a corresponding elevation 
on the slope below Burimar certainly not less than 500 
yards. It follows that since the ballistai and katapeltai 
forming the Greek artillery of that period could throw 
stones and darts a distance of only some 300 yards, and 
slingers and bowmen their missiles not much farther, it was 



CH. xx ATTACK BY CONSTRUCTING MOUND 147 

necessary to advance the position from which their 'fire* 
was to be discharged. This could be done here with effect 
only in a horizontal direction, for a descent into the ravine 
would not have increased the chance of commanding the 
higher slopes. 

The ingenious expedient of constructing a mound to 
secure this object is thus fully accounted for by the con- 
figuration of the Burimar ravine. Similarly, the use made 
of timber for its construction, whether in the form of stakes 
or tree-trunks, fully agrees with the abundance of tree 
growth that can still be seen on the slopes both above and 
below the Burimar plateau. Undoubtedly this abundance 
of timber available on the spot would supply the handiest 
material for the purpose. That the mound should have been 
advanced, as is recorded, a stadion or about 200 yards on 
the first day is readily understood in view of the compara- 
tively easy nature of the slope near the eastern edge of the 
Burimar plateau. But it becomes steadily steeper and 
steeper as the bottom of the ravine is approached, and in 
consequence the rate of the daily advance necessarily de- 
creased in proportion to the greater depth to be filled up. 
This explains why, even when on the fourth day a few Mace- 
donians had forced their way to a small hillock on the 
opposite slope, it was necessary, as Arrian tells us, to con- 
tinue work on the mound in order to join the two. 

I believe that we may safely identify this 'small hillock' 
with the shoulder of Mashlun, described above. Its level 
as measured by aneroid is about 450 feet above that of the 
bottom of the Biirimar-kandao, and about the same above 
the flat portion of Pir-sar. It is true that Arrian calls this 
small hill 'level with the Rock'. But this is easily under- 
stood, considering that a continuous slope passing Bar-sar 
connects Mashlun with the plateau portion of Pir-sar. That 
there still rose a steep height above the 'small hillock' is 

L 2 



148 AORNOS LOCATED ON PlR-SAR CH. xx 

made perfectly clear by Arrian's own narrative, where he 
describes the stiff climb which brought Alexander and his 
seven hundred to the top of the 'Rock* after the mound had 
been joined to the hillock and while the defenders were 
abandoning Aornos. 

I myself retain a very vivid recollection of the trying 
scramble over steep crags by which I gained the summit of 
Bar-sar after visiting Mashlun. I can hence realize what 
this ascent of about 350 feet must have meant for men en- 
cumbered by armour. That the height of Bar-sar was a 
very convenient place for the Macedonians to assemble, and 
then at a preconcerted signal to turn upon the retreating 
barbarians, as related by Arrian, is obvious. It is also easy 
to understand that some of the latter in their panic-stricken 
flight during the night lost their lives by falling down preci- 
pices below Pir-sar. 



CHAPTER XXI 

ANCIENT REMAINS AT PlR-SAR AND THE NAME OF 
MOUNT U#A 

WE have now seen how closely all the topographical details 
of Pir-sar agree with what our extant records tell us of 
Aornos and Alexander's operations against it. But anti- 
quarian and philological evidence may be adduced in 
further support of this identification. There is no sugges- 
tion whatsoever in our texts that the natural defences of 
Aornos had been strengthened by the hand of man, and 
we may attach all the more significance to this negative 
fact in view of the obvious desire of our authors to empha- 
size the greatness of the difficulties overcome before the 
stronghold was captured. That Aornos was recognized by 
them to have been a purely natural stronghold is clearly 
shown by the fact that they ordinarily designate it simply 
by the term petra, 'the Rock'. But we are told by Arrian 
that Alexander after the capture built there a fortified post 
and entrusted the charge of it to Sisikottos, an Indian 
deserter, who had joined him in Baktra and proved trust- 
worthy. Curtius, too, mentions Sisicostus as having been 
charged with the defence of the 'Rock' and adjoining terri- 
tory. Curtius further mentions that Alexander erected 
altars on the 'Rock' to Minerva and Victory, while Arrian 
refers merely to sacrifices performed there by him. 

In view of Arrian's statement, it is of distinct interest 
that I found the ruinous remains of what undoubtedly was 
a small fort on the summit of Bar-sar (Fig. 86). The 
walls occupy all the level space there is on the top, and to 
the north, towards Lande-sar, descend also on the slope. 
They form an irregular quadrilateral, of which the longest 
side eastwards measures 136 feet and the shortest to the 
north sixty feet. The walls, five feet thick throughout, are 



ISO ANCIENT REMAINS AT PlR-SAR CH. xxi 

deeply buried in debris and earth, largely humus deposited 
by decay of the luxuriant forest vegetation that has grown 
up and flourished, evidently for centuries, among and over 
the ruins. It was only by a careful search that I was able 
to trace the lines of the enclosing walls and some small 
rooms in the southern part of the enclosed area. Such ex- 
cavation as was possible in the time and with the labour 
available showed masonry of a type not unlike that found 
at Bir-kot and at ancient dwellings of early Buddhist times 
in Swat, stone slabs, unhewn but fairly uniform in thick- 
ness, being set in mud plaster. Among the potsherds dis- 
covered on the floor of one of the rooms there were some 
showing ornamentation similar to that found at Buddhist 
sites of Swat but less finished. 

What pointed to considerable antiquity was the far- 
advanced decay of the whole structure, as compared with 
the fair condition in which most of the ruined dwellings and 
fortified mansions dating from Buddhist times are found at 
Swat sites. Yet these, by their position, are far more 
exposed to erosion and other destructive factors than the 
very top of Bar-sar could be. The position is such as could 
not have been chosen for any other purpose than defence. 
Whether the remains in question can go back as far as the 
Macedonian invasion, and whether they mark the spot 
where the fort erected under Alexander's orders may have 
stood, are questions that it is impossible to answer without 
thorough investigation, such as was not possible at the time 
of my visit. But it is certainly noteworthy that the ruined 
fort crowns just that height which protects the Pir-sar 
plateau on the side where, as we have seen, it was most 
exposed to attack. 

The old Gujars, who had been summoned from the ham- 
lets below as depositories of local lore (Fig. 88), knew of 
no special tradition attaching to these ruined walls. Among 




1III.11S M.\K MIDIHL OF IMR-SAR RTDC.E, WITH BAR-SAR AND LANDE-SAR 

IN DISlANCh. 




CH. xxi TRADITIONS OF GUJARS 151 

them was Ibrahim Baba, a venerable old man, who was 
brought up with much trouble in a litter, and was declared 
to be a fountain-head of local information. He remembered 
having fought, as a young man between twenty and thirty 
years of age, against the British at the Ambela Pass in 
1862. Nor had they ever heard of Alexander having 
visited these parts. But they had been told by their elders 
that Pir-sar had served as the summer residence of a Raja 
called Sirkap, who otherwise lived below at the village of 
Sarkul on the Indus opposite Thakot. This name of 'Raja 
Sirkap' is widely attached to ancient sites in these parts on 
either side of the Indus, e.g. to the ruins of the earliest city 
at Taxila that has so far been explored. But it gives no 
clue beyond indicating a traditional belief that the Pir-sar 
plateau was occupied in early times long before the advent 
of Islam. The same Gujar informants derived the name 
Pir-sar from a saint called Pir Beghan, who is said to have 
lived on the plateau before the Pathans took the land, and 
to have been buried as a saint at the Ziarat, near the centre 
of Pir-sar. 

Whether any datable remains are concealed in parts of 
Pir-sar now under cultivation or occupied by Gujar huts 
and graveyards, it is impossible to say. But in the mosque 
that lies some 300 yards south of the Ziarat there are two 
large carved slabs of white calcareous stone, now used to 
support the roof, but undoubtedly ancient. Their exposed 
portions measure some six feet in height. They were said 
to have been dug up some time ago somewhere near the 
centre of the area. But nobody could or would indicate 
the exact spot; my inquiry here, as elsewhere, suggested, 
no doubt, an intention to hunt for buried 'treasure'. 

There still remains the philological evidence to be set 
forth. It is furnished by the name Ona, in Pashtu also 
spelt Onra, applied to the peak rising immediately above 



152 ANCIENT REMAINS AT PlR-SAR CH. xxi 

Pir-sar and also to the whole massif. We do not know the 
exact indigenous form of the local name which the Greek 
"Aopros was intended to reproduce. But if we assume 
it to have sounded *Avama> it is as easy to account for its 
phonetic transition into modern Una (Unra) as it is to prove 
that Ao/>vos was the most likely Greek rendering of it. That 
the name rendered by "Kopvos also appealed to Greek ears 
by its apparent Greek meaning '[the mountain] where 
there are no birds', is likely enough. We know from the 
reproductions of other Indian local names how ready Alex- 
ander and his companions were to seek an echo of Greek 
words in the Indian appellations that they heard. But 
there is not the least reason to doubt that Aornos was 
meant to render a genuine local name and was not a freely 
invented Greek designation. That the name Una has 
a wider local application can safely be inferred from the 
fact that the appellation Una-sar, 'head of Uija', and not 
merely Una, is used for the highest portion of the massif. 

There is definite philological evidence to show that in the 
modern name Una (Unra) pronounced with that peculiar 
cerebral n sound which in Pashtu spelling also figures as nr, 
we may safely recognize a direct phonetic derivative of an 
earlier form *Avarna, the assumed original of Aornos. I 
have shown elsewhere that the changes involved are such as 
unquestionably occur in the phonetic development of both 
Indo-Aryan and Dardic languages. 

I have left to the last the discussion of a classical notice 
which, if it is taken to refer to Aornos, as I believe it must 
be, is of quasi-chronological interest and indirectly helps to 
support the proposed location of that stronghold. Chares 
of Mytilene, one of Alexander's chief officials, is quoted by 
Athenaeus as having recorded in his history of Alexander 
a method of conserving snow used at the siege of the Indian 
'town of Petra'. According to Chares, we are told, 'Alex- 



CH. xxi EVIDENCE OF NAME OtfA 153 

ander ordered thirty trenches to be dug close to each other 
and to be filled with snow, branches of trees being also 
thrown in, in order that the snow in this way may be pre- 
served longer'. I believe that in this stray notice we have 
a useful indication both of the elevation of the ' Rock' and of 
the season when Alexander besieged it. 

We know from a record of Aristobulos, who shared 
Alexander's campaign and is quoted by Strabo, that the 
army, having set out for India from the Paropamisadai, 
i.e. the valleys between the Hindukush and Kabul, after 
the fall of the Pleiades, spent the winter in the hill territories 
of the Aspasioi and Assakenoi, but in the early spring 
descended to the plains and moved to Taxila, and thence 
to the Hydaspes and the country of Poros. That the siege 
of Aornos was the last of the major operations carried out 
before the crossing of the Indus and the advance to Taxila 
is quite certain from the concordant records of Arrian and 
the other historians. It is also certain that this operation 
was undertaken after Alexander had descended to the plain 
of the Peshawar valley. We can therefore place that siege 
neither much before nor much after the month of April 
326 B.C. 

Now, from my personal experience on this journey and 
from the climatic conditions previously observed in similar 
localities on the North-West Frontier, I may safely assert 
that in April snow could not be found there much below an 
elevation of 6,000 feet. On the other hand should water be 
needed for large numbers, the need of preserving snow for 
drinking purposes on heights situated between 6,000 and 
9,000 feet might well arise at a season when the slopes are 
exposed to the powerful sun of an Indian spring. From 
what I saw on my way past the Uiia peak and the adjacent 
heights, I believe that the expedient recorded by Chares 
would probably nowadays also recommend itself if troops 



154 ANCIENT REMAINS AT PlR-SAR CH. xxi 

were obliged for a time to occupy that high ground and 
its southern slopes. The spring of the year referred to in 
this narrative had been quite exceptionally belated. Yet 
at the time of my visit at the very end of April we found 
snow only in small sheltered hollows on the northern slopes 
of Mount Oija and none at all on the south. The fine 
spring a&ove 'Little Uoa' and another at Adramar, about 
the same distance on the opposite side of the peak, would 
scarcely suffice for a large force encamped on this part of 
the range. It would therefore be no more than an act of 
prudence if a commander, faced by uncertainty as to the 
length of his stay on those heights, took steps to conserve 
whatever remained of the winter's snowfall. We thus see 
that this fragmentary reference also perfectly accords with 
that combined evidence of texts, topography, and name 
which has led us to locate Aornos on the rock-girt site 
adjoining Mount Oiia. 



CHAPTER XXII 

FAREWELL TO AN HISTORIC SITE AND ITS STORY 

WITH so much to examine on the ground and so much to 
reconstruct of a great episode in the distant past, I had 
more than enough to occupy my eyes and thoughts during 
those strenuous but happy days on the heights of Pir-sar. 
The longer I studied the ground, with all its formidable 
obstacles to the movement and even to the mere mainte- 
nance of large numbers of men, the more amazed I felt at the 
unmeasured energy and ambition which led Alexander to 
attempt the conquest of so inaccessible a mountain fastness. 
No less surprising appeared the devotion and unremitting 
endurance of his Macedonians. The mere thought of the 
vast distances that they had covered since they set out to 
follow their young king from the shores of the Aegean 
would have sufficed to cast glamour over the actual scene 
of one of their many achievements. 

The record of this particularly famous exploit, as it has 
come down to us, might well have seemed, before the scene 
of it was known, to bear an air rather of romance than 
of history. Yet now face to face with the tremendous diffi- 
culties that nature itself had here opposed to the invaders, 
I could only wonder that the story of Aornos should have 
escaped being treated altogether as a mythos. But then 
the whole tale of Alexander's triumphant achievements 
from the Mediterranean far into Central Asia and India is 
full of incidents testifying to such combined energy, skill, 
and boldness as would be sought rather in a divine hero of 
legend than in a mortal leader of men. 

During the three days of my stay, there was plenty of 
life on the Pir-sar plateau. Khans from the large villages 
below on the Indus came to visit us, each a petty feudal 
baron attended by his armed retainers, while strings of 



156 FAREWELL TO AN HISTORIC SITE CH. xxn 

Gujars toiled up with supplies. The Badshah's relations with 
Baradar Khan of Thakot, the influential chief of the clans 
across the river, were still strained, and there was good 
reason to fear that the latter might tamper with the alle- 
giance, only recently secured, of the tribes on the right bank. 
So a nuipber of forts had been lately built along the river 
to ward off inroads from that side, and the commanders of 
these, too, did not fail to put in their appearance. The 
Sipah-salar and all his host bore bravely with exposure and 
cold. But there was no attempt to disguise their relief 
when the completion of my task made it possible to fix the 
date for our departure. 

Had I not been obliged to consider the discomforts to 
which the men were subjected, I should gladly have re- 
mained longer on this classical counterpart of 'my' Kashmir 
alp, so glorious were the views it offered. To the north rose 
in a long line the great snowy peaks above Kana and 
Torwal (Fig. 80), with their continuation towards parts of the 
Indus Kohistan that are as yet unexplored. Far away in the 
distance Shah 'Alam's keen eyes recognized the big spur 
over which a pass leads from Tanglr to his home of exile 
in Kandia. It cannot be far from the big bend of the Indus 
where it turns in a narrow canon to the south, and in 1913 
when I crossed from Darel to Tanglr my eyes must already 
have rested on this spur. 

To the north-east of the Indus my view extended to 
another personal link with the past; for in the chain of 
snow-covered mountains that divide the Indus Kohistan 
from Kaghan there was more than one peak that I must 
have sighted on evening climbs above my Battakundl 
camp of the summers of 1904 and 1905. Farther down 
I could see winding stretches of the greenish waters of the 
Indus, over 5,000 feet below Plr-sar (Fig. 87). Beyond to the 
east the eye wandered over the open valleys of Nandihar 




Sj INDUS RIYI R WII1I SXOW-C OVI RTI) RAN'GL '1O\\ARDS KAGHAN, SEEN FROM 

HI LOU KL Z-S VR 




CH. xxii DISTANT VISTAS FROM PlR-SAR 157 

and Pakhll far away to the Murree hills, still glittering here 
and there with the snow of this abnormally late spring. The 
distances were far too great to recognize there any of the 
small hill stations where British troops had already taken 
up their summer quarters to escape the heat of the plains. 
But the big massif of the Black Mountains hiding Agror 
and Hazara seemed quite a close neighbour across the 
Indus valley. And as the eye followed the river's tortuous 
course to the south I could clearly see through my glasses 
how useless it would be to look for another plateau like 
Pir-sar among the jumble of chopped and serrated hills 
stretching down through Puran and the Chagarzai country 
towards Buner and Mahaban. 

It was in that direction that my next moves were to take 
me. On general grounds it is probable that some of these 
hill tracts between Buner and the Indus had also been 
visited by Macedonian forces after Aornos had been cap- 
tured. But the notices left to us of Alexander's movements 
immediately after that great feat are too brief and divergent 
in their details to permit us to trace his route with any cer- 
tainty on the ground. So I may conveniently sum up here 
what little we are told of the remainder of Alexander's 
campaign before he crossed the Indus to start on the in- 
vasion of the Panjab. 

Arrian (Anabasis, iv. xxx. 5) tells us that Alexander 
moved from the 'Rock' into the territory of the Assakenoi, 
having been informed that the brother of Assakenos, with 
elephants and a host of neighbouring barbarians, had taken 
refuge in the mountains of that region. When he reached 
the town of Dyrta in this territory, he found it, together 
with the surrounding district, abandoned by its inhabitants. 
Thereupon he detached certain commanders to examine the 
localities and to secure information from any barbarians 
captured, particularly about the elephants. We have seen 



158 FAREWELL TO AN HISTORIC SITE CH. xxii 

above that Assak^nos was the ruler whose capital Massaga 
was taken on the Macedonians' first entry into Lower Swat. 
Hence the mountain region in which his brother had taken 
refuge, and which was reckoned as pan of the territory of 
the Assakenoi, might well have been Buner ; for this, as the 
records of the Chinese pilgrims clearly show, was in ancient 
times included in Swat territory, just as it is now once more. 
But the position of Dyrta has not been identified, and no 
other indications are furnished. 

To Buner, however, seems to point what we are next 
told about Alexander's march on the Indus: 'and the army 
going on before made a road for him, as those parts would 
otherwise have been impassable/ This description would 
well apply, as first suggested by General Abbott, to the 
most direct route leading from the central parts of Buner 
to the Indus along the Barandu river; for the lower valley 
of the latter, as yet unsurveyed and in part inaccessible 
owing to the colony of 'Hindustani fanatics' at present 
settled there, is reported to be a narrow gorge in places 
impracticable for traffic. 

From captives Alexander learned that the Indians of that 
territory had fled to Abisares, i. e. to the ruler of Hazara. 
They had left the elephants behind by the river, and these 
Alexander succeeded in capturing. Finally we are told 
that, serviceable timber having been found by the river, 
this was cut by the troops and the ships built with it taken 
down the Indus to where a bridge had long before been 
constructed by the other portion of the army. 

Diodorus' account of what followed the capture of Aornos 
is very brief. We are told by him that Aphrikes, an Indian 
chief, was hovering in that neighbourhood with twenty 
thousand soldiers and fifteen elephants. The chief was 
killed by his own men, who brought his head to Alexander 
and thereby purchased their own safety. The elephants 



CH. xxii ALEXANDER'S SUBSEQUENT OPERATIONS 159 

wandering about the country were secured by the king, who 
then arrived at the Indus, and finding it bridged gave his 
army a rest of thirty days before crossing to the left bank. 

Curtius' account, evidently taken from the same source, 
supplements the above by some details, which, however, do 
not furnish any clear topographical guidance. Alexander 
is said to have marched from the ' Rock' to Ecbolima, which 
evidently is the same place as Arrian's Embolima. Having 
learned that a defile on the route was occupied by twenty 
thousand armed men under Erix, he hurried forward, 
dislodged the enemy with his archers and slingers, and 
thus cleared a passage for his heavy-armed troops behind. 
Erix was killed in flight by his own men and his head 
brought to Alexander. Thence he arrived after the sixteenth 
encampment at the Indus, where he found everything pre- 
pared by Hephaistion for the crossing. 

Here, at the starting-point of his invasion of India proper, 
we must leave the great conqueror. Alexander's triumphal 
progress through the wide plains of the Panjab has, owing 
to the fascination exercised at all times by the wonders of 
distant India, attracted the chief interest of his historians, 
ancient as well as modern. But those who are familiar with 
the natural difficulties of the territories beyond the present 
North-West Frontier and with their military history in 
recent times may well be even more impressed by the great- 
ness of the obstacles overcome by Alexander's genius and 
the pluck and endurance of his hardy Macedonians in the 
course of the long campaign that preceded the invasion. 



CHAPTER XXIII 

THROUGH CHAKfiSAR AND PORAN 

IT was with regret that I took leave of Aornos on the morn- 
ing of April 30th. I was therefore glad that the long and 
ratherfatiguing march, which brought us down to Chakesar 
in the evening, lay for the most part along the Upal range 
and allowed me to look back again and again on that his- 
toric scene ; so conspicuous is the position of Pir-sar on the 
flank of dominating Ooa-sar. 

At Chakesar, the chief place of the valleys held by the 
Azi-khel tribe, I felt brought back to the present and to a 
recent and somewhat turbulent past. The two days that 
I halted there were pleasant and interesting enough in their 
own way. On the hill-sides around many ruinous remains 
were to be seen of walls marking sites of ancient dwelling- 
places. They evidently had been quarried for centuries to 
supply stones for the small town and for the supporting 
walls of cultivation terraces. During the years before the 
Badshah's rule was established, Chakesar had seen a good 
deal of fighting between the chief local families. But now 
their defensive towers had all been razed to the ground and 
a strong fort built for the Badshah's 'Hakim'. 

The exposure and fatigue to which the men had been 
subjected during those happy days on the height of Aornos 
and the marches to and from it obliged me to make a two 
days' halt at Chakesar. It felt warm enough down there 
at an elevation of a little less than 4,000 feet, and flies and 
mosquitoes were celebrating spring revels all round my 
tent. Nevertheless I welcomed the short halt, for I had 
much writing work to do. It also enabled me to record a 
hitherto unknown Kohistam tongue, spoken only by some 
sixty to eighty households of the small Dard community of 
Batera higher up in the gorges of the Indus. Even polyglot 




89. AT THE 'MIDDLE MOSQUE' OF CHAKESAR. 



CH. xxiii A SEAT OF ISLAMIC LEARNING 161 

Raja Shah 'Alam, who had first told me about it, found this 
form of Kohistam speech difficult to interpret. 

It happened, fortunately, that, besides some men of 
Batera whom saeva paupertas due to the extreme scarcity 
of cultivable land in their own country had driven to seek 
work with Pathan Maliks near Chakesar, two intelligent 
visitors from Batera had just then been drawn to Chakesar 
by the fame of a pious Mian teacher. These all spoke 
Duberi as well as Pashtu in addition to their own 
tongue. So the taking down of linguistic specimens of 
'Batochr for Sir George Grierson's learned analysis did 
not prove quite so troublesome a business as I had at first 
expected. 

Chakesar lies close to the meeting-place of four much- 
frequented routes from Kana, upper Ghorband, Puran, and 
the Indus valley. There is plenty of good cultivable ground 
held by the virile AzI-khel clan in the neighbourhood and 
in the valleys through which those routes lead. These 
facts account for the importance of Chakesar not merely as 
a local trade centre but also as a seat of Muhammadan 
theological learning. The visits I paid one evening and 
again on the morning of my departure to the four 'Schools' 
of Jumats, established in the shady courts of different 
Mosques (Fig. 44), was quite an interesting experience. 
Half a dozen 'big Mullahs', headed until a year or two ago 
by Khan-khelo 'Abdul Jalll, a famous expounder of the 
Law belonging to one of the chief Azi-khel families, are 
drawing students from all parts of the Frontier region, and 
also from beyond it, to this quaint semblance of a small 
medieval University. Since 'Abdul Jalil's death the fame 
of the Chakesar schools was said to have become somewhat 
dimmed. Yet over a hundred and twenty 'Talib-ilms' were 
still to be counted, all of them as well as their teachers 
maintained in food and quarters by the Pathan landowners 



162 THROUGH CHAKESAR AND PflRAN CH. xxm 

of the place, with some contribution from the tithe now 
levied by the Badshah. 

The amenities of Chakesar, in the way of good rations, 
helpful local patrons, and an equable climate, are evidently 
appreciated by wandering students even from a distance. 
Quite t number of the 'undergraduates' imbibing here in- 
struction in Muhammadan law and religious tradition, 
besides Arabic grammar and a modicum of Persian belles 
lettres, were said to have their homes in different parts of 
Afghanistan. When I inquired after students with whom 
I might hold converse in familiar Turk! speech, a jolly- 
looking greybeard was produced with evident pride, a 
product of far-famed Bukhara. He was at first much 
embarrassed when I addressed him in TurkI; for he had 
left Central- Asian parts some eleven years before, and even 
there had been accustomed to use mainly Persian, the 
speech of learned men. But gradually TurkI came back 
to his tongue. For fully twenty-seven years he had led 
the life of a wandering student and, perhaps with good 
reason, thought himself still far from that standard of 
knowledge which would allow him to settle down as a guide 
to others in matters of the holy law. 

Another aspect of the Talib-ilms' life was presented by 
a ruddy-faced Ozbeg youth, Maqdiim by name, whose 
acquaintance I made at the 'Middle Mosque' on the morn- 
ing of my departure. His home was at Talikhan, in 
Afghan Turkestan. He had studied at Kabul and Jalala- 
bad, and now for a year past had found instruction and 
sustenance to his taste at Chakesar. He was, of course, 
pleased to hear himself called up in his TurkI tongue, as 
he sat in a group with other students under the big Chinars 
that shade the paved court in front of the praying-hall, 
with its well-carved wooden arcade (Fig. 89). But for 
some little time neither in TurkI nor in any other tongue 



CH. xxin STUDENTS FROM TURKESTAN 163 

would speech come from his lips, though they kept steadily 
moving. 

At last the laughter of those in the crowd watching the 
one-sided dialogue brought the explanation. The young 
fellow's throat and mouth were choked with the rice he 
had hastily crammed in before rising from the big platter 
containing the morning meal that he shared with five other 
students. Eating meant for all of them competition in 
quickness of absorption. As Maqdiim was aware of the 
keen appetites of the rest, he had taken special care to store 
away what he considered his due share, before abandoning 
the treat to them. When he had at last struggled through 
these embarrassing additional mouthfuls, we had quite a 
cheery talk together about his past studies and his hope of 
soon continuing them at Delhi. The modest viaticum I 
offered him was evidently accepted as ample compensation 
for the portion of the meal he might have missed. 

But not all the frequenters of the Chakesar seat of learn- 
ing could be trusted to take so kindly to infidel strangers as 
did this genial Turki, coming from what by contrast with 
India seemed to me like a bit of Eastern Europe. Talib- 
ilms are not without reason considered among the more 
fanatical elements on the Frontier. I could therefore guess 
the reason why 'Abdul Jalil Khan, one of the local head- 
men, a fine upstanding person (Fig. 91), when guiding me 
through the narrow lanes of Chakesar, carefully took out 
his Mauser pistol from its case and in a nonchalant yet 
significant fashion carried it on his right arm. With polite 
if needless consideration for what might be my supposed 
feelings, he declared that it was a precaution he thought 
advisable on his own account, in view of some unsettled 
blood-feud. Fortunately there was no occasion for him to 
show what the automatic discharge of the bullets at close 
quarters might mean. [Since these lines were written news 

M 2 



164 THROUGH CHAKESAR AND PflRAN CH. xxm 

has reached me that my jovial guide has fallen a victim to 
vendetta.] 

Our march on May 2nd led us across the Kaghlun pass 
westwards into the wide and fertile valley of Puran. The 
picturesque side Nullah through which the ascent lay was 
clothed with luxuriant vegetation, nourished by a lively 
little stream cascading amidst moss and fern-covered rocks. 
With big clematis in full bloom and abundance of flowering 
creepers amidst evergreen shrubs, the scene seemed as if 
translated from some favoured Mediterranean region. The 
top of the ridge which the pass crosses, at about 6,200 feet 
above sea-level, offered a striking view towards both the 
head of the Ghorband valley and the high serrated water- 
shed range, crowned by the bold peaks of Dwasare and 
Ham, which divided us from Swat. Rhododendron-trees in 
full bloom furnished the escort with glowing crimson nose- 
gays, which they stuck into the muzzles of their rifles. But 
clouds soon began to shroud the heads of the peaks, and 
the descent to Aloch, the chief place of Puran, with its 
newly built fort, was made under a steady drizzle. The 
tents came in late and were pitched in pouring rain. But 
the long wait under otherwise depressing conditions was 
brightened by the opportune arrival of a heavy mail-bag 
full of cheering letters. No fewer than three of them came 
from dear friends dwelling on the flower-decked heights 
above spring-time Florence. 

Two pleasant marches took us thence through fertile 
Puran and the adjoining Mukhozai hill tract. These like- 
wise had plenty of fine scenery to offer, never yet seen by 
European eyes. At Chauga village (Fig. 92), near where 
the streams descend from the eastern slopes of rugged 
Mount Dwasare into Puran, we halted for a night. It was 
here that the Badshah's invading force, commanded by 
Ahmad 'All, his Sipah-salar and now my protector, had 



CH. xxin RECENT FIGHTING WITH CLANS 165 

four years before been hemmed in and held fast, for seven 
long days of intermittent fighting, by the confederated 
clans of Piiran, Chakesar, Ghorband, and Kana. With 
justifiable pride the alert Commander-in-chief pointed out 
to me the steep rocky spurs on either side of the valley from 
which he at last succeeded in driving the greatly superior 
force of the besiegers. 

Their hold had slackened, probably owing to the failure 
of supplies from their homes, the usual source of weakness 
of Pathan tribal gatherings of this kind. Mutual distrust 
and a want of capacity for common organization, character- 
istic of such democratic communities on the Frontier, would 
never allow the clansmen to make systematic arrangements 
of their own for 'supply and transport*. The importance of 
these had been duly recognized by the Badshah, and the 
mule corps that he maintains, together with the bridle-roads 
that he has built to these valleys as well as into other parts 
of his dominion, must go a long way towards ensuring con- 
trol over any tribal risings. In Chauga there were still 
numbers of houses standing roofless or with badly battered 
walls. But I noticed no sign of lingering resentment among 
the local Babuzai clansmen; like the rest of the Badshah's 
new subjects, they seemed to have accepted with relief the 
internal peace imposed by his regime at any rate as a 
temporary change. 



CHAPTER XXIV 

TO BUNER AND MOUNT ILAM 

ON May 5th we ascended by a good riding-road the beauti- 
ful and fertile main valley of Mukhozai westwards. The 
well-cujtivated fields where the wheat was already ripening 
betokened the advantages derived here from the absence of 
the wesk system, which is unknown to the local clan. 
Lively streams descend from the heights of Dwasare, well 
over 10,000 feet above the sea, and the fine flowering shrubs 
on the lower slopes, with much forest higher up, seemed to 
indicate an adequate rainfall. Large shady plane-trees on 
green meadowland by the streams recalled Kashmir. 

Then, on the Nawe-ghakhe pass, close on 7,000 feet 
in height, we reached the north-eastern border of the big 
territory of Buner. From that high ridge an impressively 
wide view opened over the greater portion of Buner (Fig. 
93); the bold detached hills that encircle and divide its 
large valleys rose before me as if on a relief map. With 
pleasure I greeted again the well-remembered passes and 
valleys through which I had moved more than twenty- 
eight years before in the course of my 'archaeological tour 
with the Buner Field Force'. Far away to the east I could 
take a last farewell view of the far-stretching range that 
divides Ghorband and Chakesar, with the rugged height of 
Mount Uoa standing on guard over its Indus end. Near 
below on the Buner side the eye ranged over boldly serrated 
hills, all clothed in vivid metallic green, mostly of cedars 
and firs (Fig. 94). 

In the picturesque valley of Gokand where my camp stood 
for two nights, at a height of some 4,200 feet, it felt rather 
hot and close. But a day's long ride down the valley was 
rewarded by the discovery of interesting ancient remains, 
which in 1898 had remained beyond my range. Among 



CH. xxiv GOKAND VALLEY AND PACHA 167 

them was a fine Stupa found in the little side valley of 
Top-dara below the village of Bagra (Fig. 90). It proved 
remarkably well preserved under the luxuriant growth of 
vegetation, including wild fig and other trees, which had 
managed to effect a lodgement in the solid masonry of the 
hemispherical dome and its triple base. A fine spring issu- 
ing near the ruins no doubt accounted for the choice 
of the site, and in the heat of the day was doubly grateful 
to men and ponies. The season from here onwards began 
unpleasantly to remind me of the advance that the 'hot 
weather' had already made in the plains. 

On the 7th May I made my way across the Rajgalai 
pass to Pacha, a large village containing the most famous 
shrine of Buner, the tomb of the saintly hermit known 
simply by the name of Pir Baba, 'the holy old man'. On 
the way I stumbled, as it were, upon the ruins of a smaller 
Stupa, which, half smothered under the debris and refuse 
accumulations from adjoining Gujar hovels, had remained 
apparently unknown to my local informants. At Pacha, 
which I well remembered from my first visit with the Buner 
Field Force during the punitive expedition of 1898, I 
halted for two days under the walls of the Badshah's 
moated fort, partly in consequence of the keen desire of all 
my companions to pay prolonged devotions to the powerful 
saint and partly owing to heavy rain. Fortunately the 
latter gave place for a time to mere drizzling vapour and 
mist, and allowed me to examine some curious ancient 
towers and habitations at Ramanai, high up on an outlier 
of Mount Ham (Fig. 96). But it was a somewhat exhaust- 
ing experience of an aerial Turkish bath. There was no 
ignoring the fact that over the big open valleys of Buner 
the hot weather had begun to set in, the elevation of Pacha 
being only about 2,500 feet. 

So I felt quite glad when on May loth I was able to start 



168 TO BUNER AND MOUNT ILAM CH. xxiv 

for the last item on my programme of exploration. It was 
the visit to the summit of Mount Ham, the most conspicuous 
landmark of both Swat proper and Buner. Legends of 
ancient date, as the account of Hsiian-tsang proves, cluster 
round the peak. It raises its conical head of precipitous 
crags to 9,200 feet, in noble isolation above the wooded 
spurs which radiate from it. The ascent, which occupied 
two days, started just above the large village of Bai. I 
remembered it well from a bitterly cold night I had spent 
near it in January 1898, bivouacking with my old friend 
General L. C. Dunsterville, then a young captain, and his 
old regiment, the xxth Punjabis. 

The first march brought us by the evening to the small 
village of Ilam-kile, nestling in alpine seclusion between the 
main peak and the rugged outlier of Alak-sar rising to the 
south of it. The stiff climb along cliffs and round deep-cut 
ravines was made pleasant by the varied zones of vegetation 
that we traversed. Lower down the eye could feast on the 
wonderful display of velvety orange-red blossoms hanging 
in clumps from the bare branches of trees in which I recog- 
nized with surprise the 'Dak' of the Western Panjab plains. 
Evidently the shelter and warmth afforded by the southern 
aspect of the slopes had allowed these trees to establish 
themselves on heights up to about 3,500 feet. Higher up, 
violets and a great variety of flowers, familiar to me but 
retaining as it were their incognito, owing to my botanical 
ignorance, were spread in profusion wherever fertile loam 
hid the chalk and granite rock. It was pleasant to think 
that this floral wealth had already attracted the notice of 
old Hsiian-tsang, fixed as his eyes mainly were on things 
sacred and spiritual. 

This is what his narrative tells us: 

Above 400 K south from M6ng-chieh-li is the Hi-lo mountain. 
The stream of the mountain valley flows west ; as you go up it east- 



CH. xxiv HSUAN-TSANG'S HI-LO MOUNTAIN 169 

ward flowers and fruits of various kinds cover the watercourse and 
climb the steeps. The peaks and precipices are hard to pass, and 
the ravines wind and curve. You may hear the sound of loud 
talking or the echo of musical strains. Square stones like couches 
made by art form an unbroken series over the gulley. It was here 
that Buddha once in a former birth gave up his life for the hearing 
of a half-stanza of doctrine. 

The identification of the 'Hi-lo mountain' with Mount 
Ham, long ago conjectured by M. Foucher, was proved to 
be right by the evidence relating to those 'square stones 
like couches' furnished by my visit to the summit. And 
though the distance indicated from Meng-chieh-li or 
Manglawar (corresponding roughly to eighty miles or so) 
is certainly wrong, our Chinese pilgrim guide was right in 
describing the stream of the mountain valley as flowing 
west. The principal drainage from the massif is down the 
valley that descends to below the Buner side of the Karakar 
pass, i. e. to the west-south-west, and it is by this that the 
easiest route, usually followed by modern Hindu pilgrims, 
leads up to the Jogian-sar end of the summit. So once 
again my patron saint's sense of topography has been 
vindicated, at least as regards the bearing. 

At more than one point of the ascent we had passed ruins 
of ancient dwellings and traces of terraced fields abandoned 
ages ago. But what I enjoyed more was the delightful 
coolness of our camp that night in the well-screened little 
basin where the fields and farms of Ham village nestle at 
an elevation of close on 6,000 feet. That they belong to 
Mians, supposed descendants of the holy 'Pir Baba/ of 
Pacha, seemed curiously to accord with the sacred character 
that had attached to the mountain in Buddhist times. 

From Ilam-kile the final climb of some 3,000 feet to the 
top of the peak was accomplished before midday of May 
nth. It led first over well- wooded slopes to a dip in the 
watershed towards the Jaosu pass, where I came in sight of 



170 TO BUNER AND MOUNT ILAM CH, xxiv 

the Swat river and the lower ends of the valleys that I had 
visited round Mingaora and Saidu. Then the climb took 
us steeply over and between big masses of much- weathered 
rock, often undercut and carved into quite fantastic shapes 
by prolonged water action. In places precipitous gullies, 
still deeply filled with snow, were crossed amidst a thick 
growth of firs and pines. 

Then at last we reached the tower-like mass of rock 
forming the main summit. It is crowned by four isolated 
crags like the pinnacles of a square church tower, and is 
difficult of access. The easternmost of these forms on its 
top a small platform, artificially enlarged with trunks of 
trees (Fig. 95). Ancient local worship, continued in its 
present Hindu guise, has located here the throne of Rama- 
chandra, an incarnation of Vishnu. Hither an annual 
pilgrimage brings many Hindu shopkeepers with their 
families from the villages of Lower Swat, as well as from 
adjacent parts of the border. Gujars, too, in pious rivalry 
claim the spot as the resting-place, of course, of a martyr or 
'Shahfd' of the true Faith. 

The distant panoramic view enjoyed from these crags 
was truly magnificent. From the great snowy ranges of 
the Swat Kohistan and of the Hindukush beyond Dir and 
the bend of the Indus it extended over the whole of Upper 
Swat and Buner right down to the plain of Yusufzai. A 
hot-weather haze lay over the more distant parts of the 
plain and hid Peshawar with the mountains of Tirah and 
the Safed-koh above Kabul. But these, too, are within 
view after rain or in the clear atmosphere of the autumn. 
It was an ideal spot from which to bid farewell to the varied 
tracts from which it had been my good fortune 'to lift the 
Purdah'. 

But Mount Ham rarely fails to wear a cap of cloud after 
midday; so I had soon reason to leave its age- worn head. 




0,0. VI kW ACROSS BUMIR FROM RAMANAI SPUR. 

Remains of ancn-nt habitations m foregi ound. 



I. 




97. HOLLOW ON TOP OF MOUNT ILAM, WITH SACRED SPRING AND 



CH. xxiv ANCIENT LOCAL WORSHIP 171 

Descending over the precipitous crags for some two hun- 
dred feet I found the small tree-girt hollow which holds the 
little spring and a string of round limpid pools fed by it. 
They are worshipped by Hindu pilgrims as sacred features 
of the site. My eyes rested with particular pleasure on the 
large flat- topped rocks stretching along the pools (Fig. 
97); for there could be no doubt that Hsiian-tsang had 
them in mind when he referred to the 'square stones like 
couches', shaped as if by the hand of man, which were to be 
seen on Mount 'Hi-lo'. 

The little hollow or saddle between the two summits is 
only some hundred and fifty yards across. So our camp, 
with the tents of the Sipah-salar, his men, &c., was closely 
packed in its romantic setting. The two nights we spent 
there were decidedly cold, with hoar-frost still on the ground 
when the sun rose. But there were big logs burning to 
warm the men, and the Sipah-salar's jovial old chef had 
quite a cosy kitchen under a large rock fantastically 
hollowed out like the upper jaw of a fossilized Saurian. 



CHAPTER XXV 

DEPARTURE FROM SWAT 

THE morning of the i3th May saw us set out for the valley 
which leads to Saidu. The very steep descent along a 
rugged buttress of the mountain led us to the little alp of 
Sarbab on its northern side, where the Badshah has estab- 
lished his summer quarters, and farther on to the fort and 
hamlet of Miana, where we camped. On the following day 
we moved down the valley past Kukrai and Batera, which 
we had already visited in March from Saidu, and regained 
the Swat capital. There a most hearty welcome awaited 
me from its ruler. 

A suite of little rooms, rather stuffy and dark, quite close 
to the Badshah's airy reception hall, was set apart for me. 
A walled-in little court in front suggested that they had 
once served for the accommodation of the ladies of the 
chief's family. There I spent a day busily occupied in 
receiving farewell visits from those who had shared my 
travels and in distributing suitable presents. The Chief 
Commissioner had been kind enough to provide them for 
me out of his own 'Toshakhana' in special recognition, as he 
put it, of the 'political' value of my tour. Both in my own 
temporary quarters and in the private apartment of the 
Badshah I had long talks with the ruler who, like an old 
friend, discoursed freely on the trials and struggles of the 
past and his plans for the future. 

On the morning of May i6th I said my heartfelt parting 
thanks to my kind host and protector, the Badshah, his 
promising heir-apparent Shahzada Jahanzeb Sahib, and 
the chief supporters of his rule. In the course of less than 
four hours the Badshah's motor-lorry carried me, along 
with Sipah-salar and Raja Shah 'Alam, on the unmetalled 
road that had been open for a year to the Malakand. The 



CH. xxv RETURN TO MALAKAND AND KASHMIR 173 

country traversed had become familiar to me during March ; 
but now there was an opportunity of collecting my varied 
impressions into one picture of rural wealth in the big main 
valley of 'Udyana'. Far sooner than I wished the enceinte 
and grim medieval-looking towers of the Malakand were 
reached, and under the hospitable roof of Mr. Metcalfe, the 
Political Agent, perched high above the pass, my Swat 
expedition came to its end. 

Gladly would I have extended it yet for months, but for 
the call of a heavy piece of literary work awaiting com- 
pletion, and also a sense of the trouble to which my kind 
host and his people had already been put by my tour. A 
hurried visit to Peshawar, now sweltering in its hot- weather 
atmosphere, was followed by a short but most enjoyable 
stay on the cool wooded heights of Nathiagali, the summer 
head-quarters of the North- West Frontier Province. There 
Colonel Keen's kind hospitality enabled me to see that ever- 
helpful friend once more, and to supplement my previous 
reports by word-of-mouth descriptions of what I had seen 
of the land, its capable master, and its people. Then a 
brief halt at Murree allowed me to acquaint Captain W. J. 
Norman, Deputy Director of the Frontier Circle of the 
Survey of India, with the mapping results achieved through 
Surveyor Torabaz Khan's devoted exertions, before I re- 
gained Kashmir. But even from the glorious height of my 
beloved alpine camp, my thoughts have since constantly 
reverted with delight to the happiest wandering that I ever 
enjoyed between the Pamirs and the Indian Ocean. 

FINIS 



INDEX 



Abbott, General, 124-5, X 5 8 

Abbottabad, 124. 

'Abdul Ghafur, 9, 18, 77 

'Abdul Hanan, 'ancient writings' 

forged by, 79 
'Abdul Jabbar Khan, 4 
'Abdul Jalil, Khfin-khSo, 161 
'Abdul Jaffl Khfin, 163-4 
'Abdul Latff Khfin, 30 
Abhisara, Sanskrit form of Abisares, 

59 
Abisares, ruler of Hazara, 58, 59, 123, 

158 

Achar alp, 145 

Acharo-sar peak, 117 

Adramar, spring at, 154 

Afraz-gul Khan, 9 

Afridi trader venerated as martyr, 
87-8 

Agrianians, the, 124, 135, 139 

Ahmad 'All Khan, Sipah-saiar, 28, 
74, 96, 102, 103, 156, 171, 172 

'Akhtar' or Id feast, 97-8 

Akhund of Swat, the, 3, 4; shrine of, 
65,67 

Alak-sar, 168 

Alexander the Great, 2 ; his invasion 
of Swat, 36, 40, 41-8, 58, 59, 60-1, 
113, 114, 120, 121-2, 153; besieges 
Massaga, 43-5 ; besieges Bazira and 
Ora, 45-8, 58-9, 60-1, 123; his 
siege and capture of Aornos, 113, 
118, 119, I2o-i, 135-9, J 4, 14*, 
142, 143-4, 145, 146, 147, 148, i53> 
154, 155 ; his subsequent campaign, 

157-9 

Alketas, 45, 58 

Allahl, 126 

A16ch, 164 

Amb, identified with Embolima, 125, 
126 

Amb, Nawab of, 65 

AmbSla campaign of 1862, 44, 151 

Ambulima, Sanskrit form of Embo- 
lima, 124 



Amir Khan of Kana, 108, no, HI 

Amluk-dara, 32-3, 49 

Aniline dyes, their effect on the art of 
rug-making, 90 

Aornos, 61, 104, 109, 113, 121, 124, 
126, 138, 148; former identifica- 
tion with Mount Mahaban dis- 
proved, 2, 3, 117, 125; Alexander's 
siege and capture of, 113, 118, 119, 
120-1, 135-9, 140, 141, 142, 143-4, 
145, 146, 147, H8, 153, 154, 155; 
name of, identical with tJna, 115- 
16, 151-2; ravine on, 118, 136, 138, 
139, 141, 142, 146; identified with 
Plr-sar, 118, 119, 143-8, i49~5, 
I5I-4, 155 

Aphrikes, Indian chief, 158 

Aristobulos, on Alexander's cam- 
paign, 153 

Arrian, 143; on the siege of Massaga, 
43-5; on the siege of Ora and 
Bazira, 45-6, 58, 59, 60-1, 123; 
his account of the siege and cap- 
ture of Aornos, 112, 119, 120-1, 

135-7, HO, 141, 144, 145, 146, 147, 
148, 149; on the subsequent cam- 
paign, 157-8 

Asharai crest. 132 

Aspasioi, the, 153 

Assakenoi, the, Alexander's opera- 
tions against, 42, 43, 58-9, 60-1, 
120, 121, 123-4, 135, 153, 157-9 

Assakenos, 157, 158 

Ataullah Khan, 67 

Athenaeus, reference to siege of 
Aornos quoted by, 152 

Attalos, 45 

Avalokitesvara, relievos of, 73 

Avarna, assumed original of Aornos, 

15* 

Aya=Azes, 47 
Ayilisa=Azilizes, 47 

Azes,47 

AzI-khel, the, 112, 113, 160 

Azilizes, 47 



176 



INDEX 



Babuzai clansmen, 165 
Bactria, Alexander in, 41. 
Badshah, the (see Mifcigul Gul-shah- 

zSda), 5, 6, 52, 66, 67, 68, 160, 172 
Bagra, 167 
Bai,i68 ' 
Bajaur, Alexander's operations in, 

41, 42, 47 

Bajira or Bayira = Beira or Bazira, 48 
Balai, spur of, 130 
Balo, 32 
Baio-gram, 63 
Bar-kana, no, in 
Bar-sar, hill of, 119, 129, 132, 143, 

148; ruins of fort on, 149-50 
Bar-Swat or Upper Swat (q.v.), 21 
Bar&dar Khan of Thakot, 156 
Barandu river, 158 
Barrage work, ancient, at T6kar- 

dara,35 

Batera, 160, 161, 172 
'BatochF, Kohistanl tongue, 160-1 
Battakundi, author's camp at, 156 
Bazira, 45, 48, 58, 60-1; identified 

with Bir-k6t, 46, 47, 59, " 
Bilkanai, 108, 109, no 
Bilkanai, Khan of, 106; residence of, 

108 

Blr, derivation of name, 47 
Blr-kot, 27, 35, 40, 59, 150; ancient 

stronghold of, 36-9, 40, 46; the site 

of Bazira, 46, 47, 59, 120 
Blr-kot village, 30, 31, 49, 64; Stupas 

near, 31, 32, 34, 35-6, 64 
Bird sanctuary near Manglawar, 80 
Black Mountain Expeditions, 3, 103, 

104, 116, 123, 126 
Black Mountains, 157 
Blood, General Sir Bindon, i 
Bodhisattvas, relievos of, 73 
Bolton, Sir Norman, i, 5, 172 
Bracelet, bronze, found at Chat, 114 
Branial, 91-4, 97-8 
Buddha, the, legends of his presence 

in Swat, 14, 27, 86; colossal relievo 

of, 77-8 
Buddha's clothes-washing, rock of, 

86,87 



Buddha's footprints on inscribed 

stone, 86, 88 
Buddhist monasteries in ancient 

Swat, 14, 15, 17, 20, 31, 35, 62 
Buddhist pilgrims in Swat, 2, 13-16, 

57, 81-2, 94 
Buddhist relievos, 17, 32, 33, 50-1, 

57, 62, 73, 77-8 
Buddhist seal, inscribed, 57 
Buddhist Stupas. See Stupas 
Biihler, Professor George, 78, 86, 88 
Buner, i, 2, 4, 5, 43, 122, 157, 158, 

166, 167 
Buner Field Force, 1898, i, 33, 166, 

167 
Burimar, alp of, 117, 118, 133, 144, 

146, 147 
Burimar-kandao, ravine of, 118, 132- 

133, 134, 146, 147 

Central-Asian expeditions, author's, 

3, 22, 67 

Chagarzai country, 157 
Chakdara, n, 43 
Chakesar, 109, 112, 113, 126, 133, 

160, 165, 166; seat of Muhamma- 

dan theological learning, 161-2 
Charat pass, 17 
Charbagh, 60, 81, 83, 84, 108 
Chares of Mytilene, reference to the 

siege of Aornos, 152, 153 
Charrai, 107 

Charsadda=Peukelaotis, 124 
Charus, 139 
Chat, 114, 116 
Chauga, 164, 165 
Chihil-dara, 89, 90, 109 
Chinese pilgrims to Swat, 13-16, 57, 

81-2, 94 

Chinese Turkestan, 7 
Chitral, 22, 83, 97 
Chitral campaign, 1895, i, n 
Chodgrto, 95, 97 

Churrai, 84, 89-91, 95, 98, 122, 124 
Coins of early periods found in Swat, 

32-3> 34, 39-40, 47, Si, 57, 81 
Court, M., 125 
Curtius, 43; on the siege of Massaga, 



INDEX 



177 



44, 45; on the siege of Bazira, 47- 
48, 58-9; his account of the siege 
and capture of Aornos, 112, 119, 
137, 138-9, 140, 141 j M9; on the 
subsequent campaign, 159 

'Dak* trees on Mount Ham, 168 
Danda-Nurdai, 130, 144, 145, 146 
Darband, author's visit to, 4 
Dard languages of Swat, 5, 47, 83, 

90, 91-2, 109, 160-1 
Dard population of Swat before 

Pathto conquest, 3, 4, 5> 4*, 63, 64, 

83 

Darfl, 13, 22, 23, 81, 82, 156 
Dargai, 10 
Deane, Colonel Sir Harold, i, 2, n, 

I7 , 50, 78, 86 
Deane, Lady, n 
Delai, no 
Demetrios, 45 
Diodorus, 121, 143 ; his account of the 

attack on Aornos, 121, 137-8, 140, 

143; the subsequent campaign, 

158-9 

Dlr, 83, 170; Nawab of, 4, 65 

Dosillo-sar, 18 

Dftst Muhammad Khan, 108 

Duber, 89, 106, 107, 109, 129; hill- 
men of, 1 06, 107-8, 109 

Duberl tongue, 109, 161 

Dunsterville, General L. C., 168 

Dwasare, Mount, 164, 166 

Dwellings, fortified, of Buddhist 
times, in Sw&t, 24-6, 150 

Dyrta, 157, 158 

Ecbolima=Embolima, 159 
Elephant rock mentioned by Hsiian- 

Tsang, 50 

Embolima, 124, 159 
Erix, 159 

Fa-hsien, 13, 82, 86 
Fakirs', 103, 113 
Firamorz Khan, 82 
Foucher, M., 169 



Gandhara (the Peshawar valley), 15, 
24, 44; Alexander in, 120, 121, 123, 
124, 153; Graeco-Buddhist art in, 

93 

Gandhara type of masonry, 114 
G&rwT-speaking hillmen, 96 
Gauri river =Guraios or Panjkora, 

42 
Geographic des Yakshas ('Geography 

of the Demons'), 89, 124 
Ghalagai, 50 

GhSrband, 122, 133, 161, 165, 166 
Ghorband pass, 81 
GhOrband river, 100, 109, 112 
GhSrband valley, 103, 104, 164 
Girft, King, 'Castle' of, 53-6, 57, 58, 

64,72 

Gdg-dara, 62 
Gokand, 166-7 

Gompertz, Major M. A. L., 8, 56 
Graeco-Buddhist art, remains of, 17, 

57, 64, 92-3 

Grierson, Sir George, 91, 161 
Gujars of Upper Swat, 72, 92, 100, 

103, 107, 113, 114, 115. "7, i*9, 

130, 131-2, 144, 150-1* J 5 6 > l6 7, 

170 

GuligrSm, 73 

Gumbat near Bir-kot, 31-2 
Gumbatuna, Stupas at, 26-7 
Guraios river =* Panjkora, or Gauri, 

42. 43 
Guratai, 26 

Haibat-gram, 19 

'Hasan, the quarters of, 62 

Hazftra district, 3, 123, 125, 129, 157, 

158 

Hazrat C AH Khan, 74 
Hellenistic art, remains of, in Swftt, 

64, 92-3> 175 
Hephaistion, 120, 159 
Herakles, said to have found Aomos 

impregnable, 121, 137 
Hi-lo mountain, identified withMount 

Bam, 1 68, 169, 171 
Hindukush, the, 2, 13, 96, 97, 170 
'Hindustani fanatics' in Buner, 158 



N 



178 



INDEX 



Hsiian-tsang and his pilgrimage 
through Swat, 2, 13, 14, 15, *8, 5, 
57, 73, 76, 77, 81, 86, 87, 88, 168-9, 
171 

Humphrys, Sir Francis, zo 

Humphrys, Lady, 10 

Hunza gorges, 75 

Hydaspes'river, 153 

Ibrahim Baba, 151 

Id feast, 97-8 

Ilam,Mount, 27-8, 30, 32, 33-4, 35, 65, 
72, 73, 106, 164, 168-71; sanctity 
of, 32, 33-4, 72; Hi-lo mountain 
identified with, 169, 171 

Dam village, 169 

Ilam-kile, 168, 169 

'Imperial Annals', Chinese, 82 

Indo-Scythian rulers, coins of, 32, 34, 

40, 47, Si, 57, 81 
Indus Kohistan, 82, 106, 109, 156 
Indus river, gorges of, 73, 82 
Irrigation work, ancient, 35 
Isl^m Akhun, 79 

Jabo-sar, 103 

Jahanglr, 82 

Jahanzeb, Shahzada, son of the Bad- 
shah, 172 

Jalala, 21 

Jalalabad, 162 

Jalkot, 22, 63 

James, Colonel E. H. S., 5, 104 

Janbil valley, 64, 74 

Jaosu pass, 72, 73, 169 

Jhelam valley, 91 

Jinki-khel dan, 109 

J6gian-sar, 169 

Jumats, 'Schools' of, at Chakesar, 
161 

Kabul river, 120 

Kafiristan, 14 

Kafir-k5t, 18 

Kaghan, 129, 156 

Kaghlun pass, 164 

Kaka-khl sept of Nowshera, 94 

KaUta, 83, 96 



Kambala (Sw&ti rugs), 89 

Kambar, 63 

Kana, 89, 105, 106, 110-12, 113, 122, 
161, 165; Khans of, no, in 

Kana river, 109, 112 

Kana valley, 103, 106, 108, 112 

Kandag, 31, 36, 46 

Kandaro-sar spur, ancient dwellings 
on, 109 

Kandia, 23, 82, 108, 129, 156 

Kanjar-kOfce, 31 

Kansu, fortified dwellings in, 25 

Karakar pass, 33, 169 

Karakar valley, 46 

'Kar-khana' (factory), the Badshah's, 
66 

Kardrai, 99, 100, 112 

Kashmir, 7, 9, 12, 173 

Katgala, 43 

Keen, Colonel W. J., 6, 7, 9, 173 

Khalel pass, 74 

Khan-khelo 'Abdul Jaffl, 161 

Kharoshthi inscription, 86 

Khazana-gat, 78 

Khoaspes river Swat river or the 
Panjkora, 48 

Kh6es river, 41 

Khotan, 79 

Khushwakt race, 22, 23, 82 

Khwaja-khel, 83, 99 

'King Gira's Castle 1 , 53-6, 57, 58, 64, 
72 

Kohistan, 63, 82, 128; Indus Kohis- 
tan, 82, 106, 109, 156; Swat Kohis- 
tan, 29, 83, 87, 89, 95, 170 

Kohistanl tongue, an unknown, 
160 

'Kohistanfe* at Mingaora, 63 

Koinos, 45, 46, 124 

Korunduke, 96 

Koshujan peak=Mankial, 97 

K6tah, 22, 23 

Kotkai pass, 81 

Krateros, 124 

Kuan-yin, 73 

Kukrai, 172 

Kunar river valley, 41 

K'un-lun, the, 85 



INDEX 



179 



Kushfin dynasty, coins of, 32, 34, 40, 

5* 
Kuz-sar, 129, 132 

Landa, 130 

Landakai spur, 21, 24, 49 

L&nde-sar, 132, 149 

Larai, 108 

Lvi, Sylvain, 89 

Lilaunai, 101, 102, 103, 105 

Little t)na, alp of, 117, 130, 154; 
probable site of Ptolemy's en- 
campment, 144, 145 

Lower Swat, 12, 14, 18-21 

Mahaban, Mount, 2, 125, 129, 158; 

former identification with Aornos 

disproved, 2, 3, 117, 125 
Mahmud, the guide, 114, 115, 118 
Mahmud of Ghazna, 58 
Maira, plateau of, 129 
Maju, 130, 132 
Malakand, the, 9, 10, 11, 24, 95, 172, 

1 73; campaigns on, 44 
Mangala-pura = Manglawar, 76 
Manglawar, 50, 60, 74, 76, 81, 169 
Mang-o-p'o = Manglawar, 76 
Mankial (Koshujan), 63, 97, 116 
Mankial valley, 95, 96 
Maqdum, the Ozbeg youth, 162-3 
Mardan, 10, 85 
Marshall, Sir John, 6 
Mashlun, 133, 146, 147-8 
Mason, Major K., 9 
Massaga, Alexander's capture of, 43- 

46, 47, 58, 6x, 158 
Mastuj, 96 
M6ng-chieh-li= Manglawar, 50, 76, 

81, 168, 169 
Metcalfe, H. A. F., 9, 10, u, 12, 30, 

i?3 

Metcalfe, Mrs., 9, u 

Mifina, 73, 74, 172 

Miangul Gul-shahzada (the Bad- 
shah), 4-5. 6, n, 22, 23, 52, 63, 
66, 68-9, 82; his rise to power in 
Upper Swat, 4-5, 28-9, 65, 102, 1 10 ; 
roads constructed by, 12, 21, 49, 



68, 70, 73-4, 75-6, 81, 94, 100, 165 ; 

other material improvements due 

to, 64, 66, 95, 172; conversations 

with, 66-7, 69, 172; dispensation 

of justice by, 69, 70 
Mianguls, the, descendants of the 

Akhund of Swat, 4, 65 
Mians, descendants of holy man of 

Pacha, 169 
Mingaora, 60, 63, 64, 68, 70, 74, 75, 

122, 170 

Mora pass, 17, 18 

'Moti', the Sipah-salar's spaniel, 85 

Mukadir Shah, 92 

Mukhozai, no, 122, 164, 166 

Murree, 173 

Naji-gram, 35 

Nal, 17 

Nandihar, 126, 156 

Nathiagali, 173 

Naushirwan Khan, 76 

Nawab of Amb and Darband, 4 

Nawagai, 32 

Nawe-ghakhe pass, 166 

Nawe-kala, 23 

Nawe-kile, no, in 

Neve, Dr. Ernest, 7 

NewDelhr, i, 7, 93 

Niya Site, 93 

Nora, assumed to be the same as 

Ora, 58-9 

Norman, Captain W. J., 173 
North-West Frontier rising, 1897, 

1,21 
Nowshera, 10, 94 

Ora, Alexander's siege of, 45, 46, 58, 
59, 60, 61, 120; probable location 
of, 58-60 

Oxus river, 14, 75, 85 

Ozbeg youth at Chakesar, 162-3 

Pacha, 167, 169 
Paitai, 85, 98 
Pakhll valley, 157 
Pakhtun Wall, R&ja, 22, 82 



i8o 



INDEX 



Palds, people from, 63 

Palosa trees, 56 

Pamirs, the, 2, 13, 14, 173 

Panjab, the, 2, 8; Alexander's in- 
vasion of, 157, 159 

Panjkdra river =GaurI or Khoaspes, 
42, 43, 44, 48, 122 

Papinl Saiyids, the, 102 

Paropamisadai, the, 153 

Path&n language, spread of, 92 

Path&n Maliks, 161 

Pathan rising, 1897, i, 21 

Pathan tribes, custom of wesh among, 
52, 102, 109-10 ; invasion of Swat 
by, 33. 83, 109, 123 

Perdikkas, 120 

Peshawar, i, 3, 8, 9, 170, 173 

Peshawar valley (Gandhara), 15, 
24, 44; Alexander in, 120, 121, 
123, 124, 153 ; Graeco-Buddhist 
art in, 93 

Peshmal, 94-6 

Petra, 'the Rock', 138, 152. See 
Aornos 

Peukelaotis, 124 

Pezal-kandao pass, 130, 145 

Phillimore, Colonel R. H., 8 

Pir Baba of Pacha, 167, 169 

Pir Beghan, 151 

Pir Khushhal Baba, shrine of, 
56-8 

Pir-sar, 104, 113, "S, "7, "8, 129; 
ravine on, 118, 132-3, 134, 146, 
147; identified with Aornos, 118, 

119, 143-8, 149-50, is 1 -^ 155; 

plateau on, 119, 131-2, 155-6; sur- 
vey of, 128-34, 1 60; Ziarat on, 

*3i> IS* 

Pirdad Kh&n, of K&na, no 
Polysperchon, 59 
Poros, 153 
Pottery, decorated, of Buddhist 

period, 38, 114 
Ptolemy, Geography of, 24 
Ptolemy, the son of Lagos, 118, 135, 

136, 140, 143, 144, 145 
Puran, 122, 157, 160, 161, 164, 165 
PushkalavatI (Peukelaotis), 124 



Ragast Nullah, 87 
'Raja Gir&'s Castle', 53-6, 57, 62 
Raja Sh&h 'Alam, see Shfth 'Alam 
'Raja Sirkap', name attached to 

ancient sites, 151 
Rajgalai pass, 167 
Ramachandra, throne of, on Mount 

Ham, 170 
Ramanai, 167 

Ramaz&n fast, the, 68, 90-1 
Ramet, 94 

Ranjit Singh, Maharaja, 65, 125 
Ravine on Pir-sar, 118, 132-3, 134, 

146, 147 

Rawalpindi, 7, 8, 9 
Records of the Western Regions (Hsiian- 

tsang), 14-15 
Relievos, Buddhist, in Swat, 17, 32, 

33, 50-1, 57, 62, 73, 77-8 
Rug-making in Swat, 89-90, 98 

Safed-koh, the, 170 

Saidu, n, 12, 53, 54, 63, 64, 65-71, 

74, 170, 172 
Saiyids, Papini, 102 
Salt Range, 8 
Sambat, 85 

Sangharamas (monasteries), 14 
Sanskrit inscriptions near Mangla- 

war, 78-9 
Sarbab, 172 
Sarkul, 143, 151 
Shah 'Alam, Raja, 22, 23, 30, 82, 91, 

92, 108, 109, 114, 156, 161, 172 
Shahi dynasty, coin of, 34 
'Shahids' (martyrs), graves of, 87-8, 

102, 170 

Shahzada Jahanzeb Sahib, 172 
Shakhorai, 78 
Shalpin, 99 
Shamelai pass, 75 
Shamelai spur, 63, 75 
Shang, 143 
Shankardar, 49 
Shankardar Stupa, 28, 49-50 
Shehra Khan, 8 
Shilkai pass, 103, 105-6 
Shinase, Stupa of, 73 



INDEX 



181 



Sikandar, Sultan (Alexander the 
Great), 115 

Sikh aggression in Swat, 65 

Sipah-salar, the. See Ahmad 'All 
Khan 

'Sirkap, Raja', 151 

Sisicostus, 149 

Sisikottos, 149 

Snow preserved by Alexander for 
drinking purposes, 152-4 

Sogdiana, Alexander in, 41 

Srinagar, 7 

Stone ammunition, Blr-kot, 38 

Strabo, account of Alexander's cam- 
paign quoted by, 153 

Stupas, Buddhist, remains of, in 
Swat, 19-20, 23, 26-7, 28, 31, 32, 
34, 35-6, 49-5; 6a , 6 4, 7*, 73, 74, 
77, 80, 81, 84 

Sung Yiin, 14, 94 

Su-p'o-su-tu (the Swat river), 15 

Survey of India, the, 6 

Suvdstu (the Swat river), 48 

Swat, i, 2, 3, 13, 15, 25, 33, 43. 5o; 
the ancient Uddiyana, 2, 13, 27, 
89; Chinese pilgrims in, 2, 13-16, 
57, 81-2, 94; legends of the Bud- 
dha's presence in, 14, 27, 86; White 
Hun conquest of, 15; Buddhist 
monasteries in, 14, 15, 17, 20,31,35, 
62; ancient dwellings of Buddhist 
times in, 24-6, 150; Pathan inva- 
sion of, 33, 83, 109, 123; Alexan- 
der's campaign in, 36, 40, 41-8, 58, 
59, 60-1, 113, 114, 120, 121-2, 153 

Swat Kohistan (or Torwal, q.v.), 83, 
87,89,95,170 

Swat, Lower, 14, 18-21 

Swat, Upper, i, 4, 21, 59-60, 63, 73, 
9o;theMiangulGul-shahzada's rise 
to power in, 4-5, 28-9, 65, 102, no 

Swat river, i, 3, 15, 42, 48, 63, 75, 94 

'Swati rugs', 89-90, 98 

Swatls of Hazara, 123 

Takhta pass, 143 
Taklamakan, the, 92-3 
Talash, 43 



Talib-ilms* of Chakesar, 161-2, 163 

Talikhan, 162 

Talun, 130 

Tangir, 22, 23, 82, 92, 97, 108, 156 

Tarim basin, Graeco-Buddhist wood- 
carvings in, 93 

Taxila, 151, 153 

Thakdt, 104, 126, 143, 151; Khan of, 
156 

Thana, 11, 12, 17, 18-19, 63, 64 

Tien-shan, the, 10 

Tirah, 170 

Tirat, 86 

Tokar-dara, 35 

Top-dara, Stupa of, 19-20, 26, 167 

Torabaz Khan, the Surveyor, 8, 9, 
18, 83, 97, loo, 109, 128, 173 

Torwal, 3, 5, 42, 63, 64, 89, 98, 122, 
129; wood-carvings in, 64, 93; 
textile products of, 89-90, 98 

TSrwall dialect, 90, 91, 92 

Tungan rebellion, Kansu, 25 

Turkestan, Chinese, Buddhist shrines 
in, 7 

Uddiyana (Swat), 2, 13, 27, 89 

Ude-gram, 49, 53,^54, 57,62; prob- 
able location of Ora, 58-60 

Udydna, Sanskrit name for Uddi- 
yana, 13, 173 

Una, Little, alp of, 117, 130, 154; 
probable site of Ptolemy's encamp- 
_ ment, 144, 145 

tlna, Mount, 115, 117, 143, 166; the 
name rendered in Greek as Aornos, 
115-16, 151-2. 

Una-sar, 117, 130, 143, 144, 160 

Upal, 112, 113, 114, 129 

Upal pass, 113, 115 

Upper Swat (Bar-Swat), i, 4, 21, 59- 
60, 63, 73, 90; the Miangul Gul- 
shahzada's rise to power in, 4-5, 
28-9, 65, 102, 1 10 

Upper Swat Canal, 10 

Ushu, 96 

Utr6t, 96 

Uttarasena, King, Stupa attributed 
to, 49, 50; relievo of, 51 



i82 INDEX 

Wauhope, Colonel R. A., 3, 104, Wu-k'ung, 15, 16, 76 

125-6 

Wazir Hazrat 'All, 66 
Wi*. Path&n custom of, 52, 102, 



s, invasion of, 15, 166 J"J5 ^ gp 65 

showing Hellenistic Yusufzai P lam ' I0 ^ I7 
influence, 63-4, 92-3 

Wu-ch e ang or Wu-chang-na (Swftt), Ziftrats:ofRrKbushh&lB&bft J s6,S7; 

X 3* 5> 57" on summit of Plr-sar, 131, 151 



PUNTED IK GEEJLT BRITAIN AT TEE UNIVERSITY PRESS, OXFORD 
BY JOHN JOHNSON, PRINTER TO THE UNIVERSITY